Smart necklace with stereo vision and onboard processing

Abstract

A wearable neck device for providing environmental awareness to a user, the wearable neck device includes a flexible tube. A first stereo pair of cameras is encased in a left portion of the flexible tube and a second stereo pair of cameras is encased in a right portion of the flexible tube. A vibration motor within the flexible tube provides haptic and audio feedback to the user. A processor in the flexible tube recognizes objects from the first stereo pair of cameras and the second stereo pair of cameras. The vibration motor provides haptic and audio feedback of the items or points of interest to the user.

Description

BACKGROUND

Field

The present disclosure relates to a wearable device, which provides haptic and audio feedback based on stereo camera input.

Description of the Related Art

Wearable cameras provide recording and documenting of a user's experience, often from the same or similar point of view or field of view (FOV) of the user. However, these devices are passive recorders, and do not provide real time processing and information about the scene in the FOV. Certain users, such as blind persons, may desire additional feedback relating to the environment. Other wearable cameras may be designed to assist blind persons. However, such devices lack stereo cameras for reliable depth perception information.

Thus, there is a need for an unobtrusive device which augments a user's environmental awareness with depth perception and object recognition.

SUMMARY

The present disclosure relates to a smart necklace which provides audio and haptic feedback based on stereo camera input. One aspect of the present disclosure is to provide a wearable device which can recognize objects for increased environmental awareness and obstacle avoidance. Another aspect of the present disclosure is to provide a wearable device which assists in navigation. Yet another aspect of the present disclosure is to provide a smart necklace for social interaction.

In one implementation, a wearable neck device for providing environmental awareness to a user comprises a flexible tube defining a cavity and having a center portion, a left portion and a right portion. A first stereo pair of cameras is positioned on the left portion of the flexible tube and a second stereo pair of cameras is positioned on the right portion of the flexible tube. A vibration unit is positioned within the cavity and configured to provide haptic and audio feedback to the user. A processor, also positioned within the cavity, is configured to receive video frames from the first stereo pair of cameras and the second stereo pair of cameras, provide object recognition of items in the video frames, identify points of interest to the user based on the object recognition, and control the vibration unit to provide haptic and audio feedback to the user based on the points of interest.

In another implementation, a wearable neck device for providing environmental awareness to a user comprises a band defining a cavity and having a center portion, a left portion and a right portion. A first stereo pair of cameras is positioned on the left portion of the band and a first camera is positioned to a side of the first stereo pair of cameras. A second stereo pair of cameras is positioned on the right portion of the band and a second camera is positioned to a side of the second stereo pair of cameras. A vibration unit, positioned within the cavity, is configured to provide haptic and audio feedback to the user. A processor, also positioned within the cavity, is configured to receive video frames from the first stereo pair of cameras, the first camera, the second stereo pair of cameras and the second camera, provide object recognition of items in the video frames, identify points of interest to the user based on the object recognition, and control the vibration unit to provide haptic and audio feedback to the user based on the points of interest.

In yet another implementation, a method of navigation using a wearable neck device comprises recognizing objects with a stereo pair of cameras of the wearable neck device, determining a location of the wearable neck device with respect to the objects, determining a route to a destination that avoids the objects, providing a first audio or haptic cue indicating the route, and providing a second audio or haptic cue when the wearable neck device reaches the destination.

BRIEF DESCRIPTION OF THE DRAWINGS

Other systems, methods, features, and advantages of the present invention will be or will become apparent to one of ordinary skill in the art upon examination of the following figures and detailed description. It is intended that all such additional systems, methods, features, and advantages be included within this description, be within the scope of the present invention, and be protected by the accompanying claims. Component parts shown in the drawings are not necessarily to scale, and may be exaggerated to better illustrate the important features of the present invention. In the drawings, like reference numerals designate like parts throughout the different views, wherein:

FIG. 1 is a block diagram of a smart necklace according to an implementation of the present disclosure;

FIG. 2 is a flowchart of an object recognition logic according to an implementation of the present disclosure;

FIG. 3A illustrates an object recognition logic applied to a visual data set according to an implementation of the present disclosure;

FIG. 3B further illustrates the object recognition logic shown in FIG. 3A;

FIG. 3C further illustrates the object recognition logic shown in FIG. 3A;

FIG. 4 is a flowchart illustrating a method of estimating a position or orientation based on slice descriptors according to an implementation of the present disclosure;

FIG. 5A is a view of a smart necklace with a “stalk” design according to an implementation of the present disclosure;

FIG. 5B is a side view of the smart necklace in FIG. 5A;

FIG. 5C is a back view of the smart necklace in FIG. 5A;

FIG. 6A is a view of a smart necklace with a “pod” design according to an implementation of the present disclosure;

FIG. 6B is an under-side view of the smart necklace in FIG. 6A;

FIG. 6C is another view of the smart necklace in FIG. 6A;

FIG. 7A is a view of a smart necklace with a “paddle” design according to an implementation of the present disclosure;

FIG. 7B is a side view of the smart necklace in FIG. 7A;

FIG. 7C is a close-up view of a paddle of the smart necklace in FIG. 7A;

FIG. 7D is a close-up view of detachable pucks of the smart necklace in FIG. 7A;

FIG. 8 is a view of a smart necklace with a “twist” design according to an implementation of the present disclosure; and

FIG. 9 is a view of a smart necklace with a removable tube having a camera according to an implementation of the present disclosure.

DETAILED DESCRIPTION

Apparatus, systems and methods that implement the implementations of the various features of the present application will now be described with reference to the drawings. The drawings and the associated descriptions are provided to illustrate some implementations of the present application and not to limit the scope of the present application. Throughout the drawings, reference numbers are re-used to indicate correspondence between referenced elements.

In one implementation, a smart necklace 100 includes an onboard processing array 110, which communicates with a sensor array 120, an interface array 130, and a component array 140. The onboard processing array 110, the sensor array 120, the interface array 130, and the component array 140 are exemplary groupings to visually organize the components of the smart necklace 100 in the block diagram of FIG. 1 and are not limiting or necessarily representative of any physical groupings. In addition, certain implementations may have more or less components shown in FIG. 1.

The onboard processing array 110 includes a processor 111, a memory 112, and a storage 113. The processor 111 may be a computer processor such as an ARM processor, DSP processor, distributed processor, or other form of central processing. The memory 112 may be a RAM or other volatile or nonvolatile memory used by the processor 111. The storage 113 may be a non-transitory memory or a data storage device, such as a hard disk drive, a solid state disk drive, a hybrid disk drive, or other appropriate data storage, and may further store machine-readable instructions, which may be loaded into the memory 112 and executed by the processor 111.

The sensor array 120 includes a stereo camera 121, a camera 122, an inertial measurement unit (IMU) 123, a global positioning system (GPS) 124, and a sensor 125. The stereo camera 121 may be a stereo camera comprising two cameras offset by a stereo distance. The stereo distance may be optimized for the two cameras. The smart necklace 100 may have more than one stereo camera 121, as will be further described below. The camera 122 may be a camera or other optical sensor not part of a stereo camera pair. The IMU 123 may be an IMU which may further comprise one or more of an accelerometer, a gyroscope, and/or a magnetometer. The GPS 124 may be one or more GPS units. The sensor 125 may be one or more sensors which provide further information about the environment in conjunction with the rest of the sensor array 120. The sensor 125 may be, for example, one or more of a temperature sensor, an air pressure sensor, a moisture or humidity sensor, a gas detector or other chemical sensor, a sound sensor, a pH sensor, a smoke detector, a metal detector, an actinometer, an altimeter, a depth gauge, a compass, a radiation sensor, a motion detector, or other sensor.

The interface array 130 includes a microphone 131, a speaker 132, a vibration unit 133, an input device 134, and a display 135. The microphone 131 may be a microphone or other device capable of receiving sounds, such as voice activation/commands or other voice actions from the user, and may be integrated with or external to the smart necklace 100. The speaker 132 may be one or more speakers or other devices capable of producing sounds and/or vibrations. The vibration unit 133 may be a vibration motor or actuator capable of providing haptic and tactile output. In certain implementations, the vibration unit 133 may also be capable of producing sounds, such that the speaker 132 and the vibration unit 133 may be the same or integrated. The input device 134 may be an input device such as a touch sensor and/or one or more buttons. For example, the input device 134 may be a touch sensor used as a slider to adjust settings as well as act as a button for making selections, similar to a touchpad. The display 135 may be a display, integrated into the smart necklace 100 or wirelessly connected to the smart necklace 100, and may be capable of displaying visual data from the stereo camera 121 and/or the camera 122. In other implementations, the display 135 may be another visual alert device, such as one or more LEDs or similar light source.

The component array 140 includes a battery 141, an antenna 142, and an input/output (I/O) port 143. The battery 141 may be a battery or other power supply capable of powering the smart necklace 100. The battery 141 may have a connection port for recharging, or may be wirelessly recharged, such as through induction charging. The antenna 142 may be one or more antennas capable of transmitting and receiving wireless communications. For example, the antenna 142 may be a Bluetooth or WiFi antenna, may be a radio frequency identification (RFID) antenna or reader, and/or a near field communication (NFC) unit. The I/O port 143 may be one or more ports for connecting additional peripherals. For example, the I/O port 143 may be a headphone jack, or may be a data port. The antenna 142 and/or the I/O port 143 allows the smart necklace 100 to connect to another device or network for data downloads, such as updates or map information or other relevant information for a particular application, and data uploads, such as status updates. Further, the antenna 142 and/or the I/O port 143 allows the smart necklace 100 to communicate with other smart necklaces 100 for distributed computing or sharing resources. The smart necklace 100 described herein is generally a stand-alone device. However, in other implementations, the smart necklace 100 may be configured or optimized to work in conjunction with other devices. For example, smartphones, tablets, or other mobile devices may wirelessly connect to the smart necklace 100 for shared resources and processing. The mobile device may act as a display unit for the smart necklace 100. The smart necklace 100 may further have specific protocols for interacting with mobile devices or other smart necklaces 100.

The smart necklace 100 is a lightweight, wearable smart device that is worn around the user's neck for environmental awareness, navigation, social interactions, and obstacle avoidance through real-time feedback. The smart necklace 100 is capable of recognizing objects around the user, in order to alert the user. For example, the smart necklace 100 may be used by a blind person to aid in environmental awareness and navigate safely around obstacles. The smart necklace 100 provides the user audio and haptic feedback through the speaker 132 and the vibration unit 133 based upon camera input from the stereo camera 121 and the camera 122.

In one implementation, the smart necklace 100 includes two pairs of stereo cameras 121, which may be positioned on either side of the user's neck. Stereo cameras provide depth information in both indoor and outdoor environments. The stereo cameras 121 may face forward, in front of a user, to establish a field of view (FOV). The stereo cameras 121 may have, for example, an FOV of around 90 degrees. The stereo cameras 121 provide 3D information such as depth in front of the user. Additional cameras 122, which may be placed to the sides of the stereo cameras 121, may increase the FOV to, for example, around 120 degrees. Alternatively, the cameras 122 may be placed where needed, such as behind the user's neck. Although the cameras 122 may be monocular, they can provide simple recognition, even without depth or distance information. For example, the cameras 122 can detect moving objects in the user's periphery. The stereo cameras 121 and the cameras 122 continuously passively recognize objects in the environment. Working in conjunction with the other sensors in the sensor array 120, the smart necklace 100 provides the user with guidance and navigation commands by way of audio and haptic feedback.

The GPS 124 provides location information, which works with the inertial guidance information, including velocity and orientation information, provided by the IMU 123 to help direct the user. The memory 112 and/or the storage 113 may store, for example, map information or data to help locate and provide navigation commands to the user. The map data may be preloaded, downloaded wirelessly through the antenna 142, or may be visually determined, such as by capturing a building map posted near a building's entrance, or built from previous encounters and recordings. The map data may be abstract, such as a network diagram with edges, or a series of coordinates with features. The map data may contain points of interest to the user, and as the user walks, the stereo cameras 121 and/or cameras 122 may passively recognize additional points of interest and update the map data. For example, the user may give a voice command, “Take me to building X in Y campus.” The smart necklace 100 may then download a relevant map if not already stored, or may navigate based on perceived images from the stereo cameras 121 and the cameras 122. As the user follows the navigation commands from the smart necklace 100, the user may walk by a coffee shop in the morning, and the smart necklace 100 would recognize the coffee shop and the time of day, along with the user's habits, and appropriately alert the user. The smart necklace 100 may verbally alert the user through the speakers 132. The user may use the input device 134 to adjust settings, which for example may control the types of alerts, what details to announce, and other parameters which may relate to object recognition or alert settings. The user may turn on or off certain features as needed.

When navigating indoors, the standalone GPS units may not provide enough information to a blind user to navigate around obstacles and reach desired locations or features. The smart necklace 100 may recognize, for instance, stairs, exits, and restrooms and appropriately store them in the memory 112 and/or the storage 113. In another example, the smart necklace 100 may determine empty seats for the user to navigate to, or may remember the user's specific seat in order to navigate away and subsequently return to the same seat. Other points of interest may be potential hazards, descriptions of surrounding structures, alternate routes, and other locations. Additional data and points of interest can be downloaded and/or uploaded to mobile devices and other devices, social networks, or the cloud, through Bluetooth or other wireless networks. With wireless connectivity, local processing can be reduced, as high level information may be available from the cloud or other remote data centers.

The smart necklace 100 may determine paths for navigation, which may be further modified for the user's needs. For example, a blind person may prefer routes that follow walls. Using the IMU 123 and/or the GPS 124 and other sensors, the smart necklace 100 can determine the user's location and orientation to guide them along the path, avoiding obstacles. The vibration unit 133 and the speaker 132 provide audio and haptic cues to help guide the user along the path. For example, the speaker 132 may play a command to move forward a specified distance. Then, special audio tones or audio patterns can play when the user is at a waypoint, and guide the user to make a turn through additional tones or audio patterns. A first tone, audio pattern or vibration can alert the user to the start of a turn, such as a single tone or a vibration from the left side of the smart necklace may indicate a left turn. A second tone, audio pattern or vibration can alert the user that the turn is complete such as two tones, or the vibration may stop, such as the left side ceases to vibrate when the turn is complete. Different tones or patterns may also signify different degrees of turns, such as a specific tone for a 45 degree turn and a specific tone for a 90 degree turn. Alternatively or in addition to tones and vibrations, the smart necklace 100 may provide verbal cues, similar to a car GPS navigation command. High level alerts may also be provided through audio feedback. For example, as the smart necklace 100 reaches a predetermined distance—such as a foot or other value which may be stored in the memory 112 and/or the storage 113 and may be adjusted—from an obstacle or hazard, the speaker 132 and/or the vibration unit 133 may provide audible alerts. As the smart necklace 100 gets closer to the obstacle, the audible alerts may increase in intensity or frequency.

The microphone 131 may provide additional environmental data, such as sounds of moving cars or other possible hazards. The microphone 131 may work in conjunction with the speaker 132, and may be placed away from the speaker 132 to prevent interference. The microphone 131 may alternatively work in conjunction with an attached audio device, such as bone conduction devices, to provide the user with audio feedback without broadcasting the audio feedback.

The smart necklace 100 may improve social interactions. For example, the smart necklace 100 may recognize faces in a room to identify potential friends, and provide the user with audio feedback identifying friends. The stereo cameras 121 and/or the camera 122 may be further able to determine additional details about persons, such as moods or expressions, or if they are engaging in physical activities, in order to alert the user. For example, the potential friend may extend a hand for a handshake or a “high five,” and the smart necklace 100 may use audio or haptic feedback to notify the user. The microphone 131 may recognize voices of other persons to identify and appropriately notify the user, or may recognize a new voice to save for future identification.

Although the smart necklace 100 is described with respect to a blind user, the smart necklace 100 may be used in other applications. For example, the smart necklace 100 may be used by peace officers and law enforcement officers as a recorder which provides additional environmental awareness. The smart necklace 100 may be further used by athletes to record sports in a real-time, first person view. For example, performing certain actions such as a swing can be recorded, including inertial motions, to analyze the motions. The smart necklace 100 may also be used in hazardous environments to provide additional safety warnings. For example, the smart necklace 100 may be used in a factory to provide a factory worker additional warning about possible hazardous conditions or obstacles. In such applications, the sensor 125 may be specifically chosen to provide particularly relevant measurements. For instance, in an environment with harmful gas, the sensor 125 may detect dangerous levels of gas and accordingly alert the user. The sensor 125 may provide low-light viewing, or the stereo cameras 121 and/or the camera 122 may be capable of night vision, to provide the user with additional environmental awareness in low-light conditions, such as outdoors at night or photo-sensitive environments. The smart necklace 100 can be a memory device to aid persons, such as Alzheimer's patients. The smart necklace 100 can aid in shopping or otherwise navigating inventories by helping to keep track of goods. The antenna 142 may be an RFID or NFC reader capable of identifying RFID or NFC tags on goods.

Referring now to FIG. 2, a flowchart of a method 200 of adjusting object detection parameters, object recognition parameters, or both object detection parameters and object recognition parameters is schematically depicted. In some implementations, the method 200 may be implemented as logic within the machine readable instructions that, when executed by the processor 111, automatically adjust object detection parameters, object recognition parameters, or both object detection parameters and object recognition parameters. It is noted that, while the method 200 depicts a specific sequence, additional implementations of the present disclosure are not limited to any particular sequence.

Referring now to FIG. 2, at block 210 the smart necklace 100 receives image data representative of the environment. As noted above, in some implementations, the smart necklace 100 is configured to acquire video or image data, which may be video frames, of the FOV of the user from the stereo cameras 121 and the cameras 122, and to then send the acquired image data of the environment to the processor 111 and/or the memory 112 and storage 113 for storage and/or processing. In some implementations, the smart necklace 100 may receive image data from a source external to the smart necklace 100, such as via the antenna 142 through a wireless network.

The image data received at block 210 may be data of a variety of forms, such as, but not limited to red-green-blue (“RGB”) data, depth image data, three dimensional (“3D”) point data, and the like. In some implementations, the smart necklace 100 may receive depth image data from an infrared sensor or other depth sensor, such as an infrared sensor or depth sensor integrated with the stereo camera 121 and/or the camera 122. In other implementations that include a depth sensor (e.g., an infrared sensor), the depth sensor may be separate from the stereo camera 121 and/or the camera 122.

Still referring to FIG. 2, at block 220, the machine readable instructions stored in the memory 112, when executed by the processor 111, cause the smart necklace 100 to detect a candidate object, with the onboard processing array 110, based on the image data received at block 210. In some implementations, the onboard processing array 110 may detect the candidate object by identifying a candidate region of the received image data, such as a region of the image that includes high entropy. For example, the onboard processing array 110 may detect a high entropy region in the acquired target image data that includes a spray bottle. In some implementations, the onboard processing array 110 may utilize a sliding window algorithm to identify the candidate region of the received image data. In implementations, the onboard processing array 110 may detect the candidate object by utilizing a feature descriptor algorithm or an image descriptor algorithm, such as scale-invariant feature transform (“SIFT”), speeded up robust feature (“SURF”), histogram of oriented gradients (“HOG”), generalized search tree (“GIST”), fast retina keypoint (“FREAK”), and binary robust invariant scalable keypoints (“BRISK”), and the like. In some implementations, the onboard processing array 110 may bias detections to one or more spatially located regions of interest based on application, scene geometry and/or prior information.

The onboard processing array 110 includes at least one object detection parameter to facilitate the detection of the candidate object. In some implementations, the at least one object detection parameter is a window size, a noise filtering parameter, an estimated amount of light, an estimated noise level, a feature descriptor parameter, an image descriptor parameter, or the like.

Still referring to FIG. 2, at block 230, the machine readable instructions stored in the memory 112, when executed by the processor 111, cause the smart necklace 100 to recognize an object, with the onboard processing array 110, based on the image data received at block 210. In some implementations, the object recognition module may recognize the object based on a candidate region identified by the onboard processing array 110.

In some implementations, the onboard processing array 110 may recognize the candidate object by utilizing a feature descriptor algorithm or an image descriptor algorithm, such as scale invariant feature transform (“SIFT”), speeded up robust feature (“SURF”), histogram of oriented gradients (“HOG”), generalized search tree (“GIST”), fast retina keypoint (“FREAK”), and binary robust invariant scalable keypoints (“BRISK”), and the like. In some implementations in which the onboard processing array 110 utilizes a feature descriptor or image descriptor algorithm, the onboard processing array 110 may extract a set of features from a candidate region identified by the onboard processing array 110. The onboard processing array 110 may then access a reference set of features of an object recognition reference model from an object recognition database stored in the memory 112 or the storage 113 and then compare the extracted set of features with the reference set of features of the object recognition reference model. For example, the onboard processing array 110 may extract a set of features from the high entropy region of the acquired target image data that includes a bottle and compare the extracted set of features to reference sets of features for one or more reference bottle models. When the extracted set of features match the reference set of features, the onboard processing array 110 may recognize an object (e.g., recognizing a bottle when the extracted set of features from the high entropy region of the acquired target image data that includes the bottle match the reference set of features for a reference bottle model). When the extracted set of features does not match the reference set of features, an object recognition error has occurred (e.g., an object recognition error indicating that no object recognition reference model matches the candidate object). When an object recognition error has occurred (e.g., referring to the example, no reference bottle model exists in the memory 112 or the storage 113), the at least one object detection parameter may be adjusted to improve the accuracy of the object detection module, as described below with reference to block 225.

In some implementations, the object recognition module may assign an identifier to the recognized object. For example, the identifier may be an object category identifier (e.g., “bottle” when the extracted set of features match the reference set of features for the “bottle category” or “cup” when the extracted set of features match the reference set of features for the “cup” object category) or a specific object instance identifier (e.g., “my bottle” when the extracted set of features match the reference set of features for the specific “my bottle” object instance or “my cup” when the extracted set of features match the reference set of features for the specific “my cup” object instance).

The onboard processing array 110 includes at least one object recognition parameter to facilitate the recognition of the object. In some implementation, the at least one object recognition parameter is a window size, a noise filtering parameter, an estimated amount of light, an estimated noise level, a feature descriptor parameter, an image descriptor parameter, or the like.

Still referring to FIG. 2, at block 240, the machine readable instructions stored in the memory 112, when executed by the processor 111, cause the smart necklace 100 to send control signals to the vibration unit 133 and/or the speaker 132 to provide appropriate haptic and audio feedback to the user. For example, if the object recognized is categorized as an obstacle, the vibration unit 133 may vibrate at an increasing rate as the smart necklace approaches it. If the object is categorized as a hazard, the speaker 132 may play a warning sound. If the object is categorized as a point of interest, the speaker 132 may play an appropriate notice, or may remain silent. As noted above, when an object recognition error has occurred, the at least one object detection parameter may be adjusted to improve the accuracy of the onboard processing array 110.

Still referring to FIG. 2, at block 225, the machine readable instructions stored in the memory 112, when executed by the processor 111, cause the smart necklace 100 to adjust at least one object detection parameter of the onboard processing array 110, with a detection tuner module, when an object recognition error has occurred. The detection tuner module may be implemented as instructions executed by the processor 111 and data stored on the memory 112 and/or the storage 113. By way of non-limiting example, in some implementations, the detection tuner module may adjust the window size utilized by the onboard processing array 110 when an object recognition error has occurred. In some implementations, the detection tuner module includes a detection tuner model and the detection tuner model adjusts the at least one object detection parameter based on the object recognition error. In some implementations, the detection tuner model maps the object recognition error to the adjusted at least one object detection parameter. In some implementations, the detection tuner model is a learned correlation model, such as a support vector machine (“SVM”) model.

Still referring to FIG. 2, at block 235, the machine readable instructions stored in the memory 112, when executed by the processor 111, cause the smart necklace 100 to adjust at least one object recognition parameter of the onboard processing array 110, with a recognition tuner module, when object recognition error has occurred. The recognition tuner module may be implemented as instructions executed by the processor 111 and data stored on the memory 112 and/or the storage 113. By way of non-limiting example, in some implementations, the recognition tuner module may adjust the window size utilized by the onboard processing array 110 when object recognition error has occurred. In some implementations, the recognition tuner module includes a recognition tuner model and the recognition tuner model adjusts the at least one object recognition parameter based on the object recognition error. In some implementations, the recognition tuner model maps the object recognition error to the adjusted at least one object recognition parameter. In some implementations, the recognition tuner model is a learned correlation model, such as a support vector machine (“SVM”) model.

Referring now to FIGS. 3A-3C, FIGS. 3A-3C present one example of a method of object recognition according to an implementation of the present disclosure. In FIG. 3A, for example, the first visual data 306 corresponds to a 2-D image of the target object 310 positioned on a plane 324 (e.g., a table). The second visual data 308 corresponds to 3-D depth data of the target object 310 positioned on the plane 324. Category object recognition is performed for analyzing, using the processor 111 and the first visual data 306. The first visual data 306 is analyzed based on a plurality of object models stored in a database, which may be stored in the storage 113. For example, the plurality of object models may include primitive shapes such as cylinders, boxes, and the like associated with corresponding parameters defining the primitive shapes. For example, the processor 111 may determine whether any portion of the first visual data 306 corresponds in shape, structure, or in terms of other pose information, to the plurality of object models stored in the database. Each of the plurality of object models may have at least one parameter. For example, an object model may be a cylinder with parameters of a height and a radius. For example, an object model may be a box with three parameters of a width, a height, and a length.

When the processor 111 searches for an object model of the plurality of object models, more than one object model may be similar in shape or structure to a portion of the first visual data 306. For example, a body of a bottle (e.g., the target object 310) may be similar in shape or structure to either a cylinder or a box. The processor 111 is configured to determine which of the plurality of object models has the closest fit for the analyzed portion of the first visual data 306. For example, the processor 111 may assign a score (for example, a recognition accuracy percentage) as to the degree of similarity between a particular object model of the plurality of object models and the analyzed portion of the first visual data 306. For example, the processor 111 may choose the object model of the plurality of object models associated with the highest associated score (e.g., recognition accuracy percentage), as the object model that corresponds to the analyzed portion of the first visual data 306. As such, in one implementation, the processor 111 determines the parameters of the chosen object model.

As described below, the plurality of object models are not fixed. The stored object models and their corresponding parameters may be supplemented or modified. In addition or in the alternative, new category object models may be learned and stored in the database based on the recognized target objects. The discussion at this juncture assumes that the method is detecting the target object 310 for the first time, and objects having similar shapes, structure, or pose information to the target object 310 as a whole are not yet encountered and stored.

Referring to FIG. 3B, an example of the category object recognition of the method is illustrated. For example, the processor 111 may examine the first visual data 306 adjacent to, around, or within the sliding enclosure 312 from left to right, starting from the top left corner of the 2-D image represented by the first visual data 306 moving right thereafter in the direction 314. The processor 111 may recognize objects within the first visual data 306 that are similar in shape or structure to an object model of the plurality of object models stored in the database. In other implementations, instead of the sliding enclosure 312, the visual data set 304, the first visual data 306, the second visual data 308, or combinations thereof may be examined as a whole to determine whether any portion of the first visual data 306 matches an object model stored in the database.

FIG. 3C illustrates a result of the category object recognition. The processor 111 may recognize that the target object 310 is similar to one of the object models. The first enclosure 350 may be a bounding box, a bounding circle, or any other shape without limiting the scope of the invention. The first enclosure 350 has a first center point 316. When the first enclosure 350 is a bounding box, the first center point 316 is the point with approximately equal distance from each side of the bounding box. When the first enclosure 350 is a bounding circle, the first center point 316 may be the center of the bounding circle. In one embodiment, the processor 111 may determine the first center point 316 such that the first center point 316 is positioned on, corresponds to, or falls within a portion of the visual data set 304 corresponding to the target object 310. The target object 310 may, for example, be positioned within, around, or adjacent to the first enclosure 350. The processor 111 determines that a first target data (which is a portion of the first visual data 306) corresponds to the target object 310 to recognize the target object 310.

Although the method described above uses a bottle as an exemplary object, the method may be used to recognize points of interest and other features, such as stairs, empty seats or buildings.

Referring now to FIG. 4, which shows a flowchart 400 of a method of estimating position or orientation based on slice descriptors. The onboard processing array 110 receives omni-directional image data representative of a panoramic FOV from the stereo cameras 121 and the cameras 122. In some implementations, the stereo cameras 121 and the cameras 122 operate within a physical environment and is configured to acquire omni-directional image data, and to then send the acquired omni-directional image data of the physical environment to the onboard processing array 110 for storage and/or processing. In some implementations, the onboard processing array 110 may receive omnidirectional image data from a source external to the smart necklace 100, such as via the antenna 142. The acquired omni-directional image data may be in the form of digital video and/or one or more digital photographs.

The onboard processing array 110 segments the omni-directional image data into a plurality of image slices. In one exemplary implementation, the received omni-directional image is segmented into eight slices (S1, S2, S3, S4, S5, S6, S7, and S8). In some implementations, the omni-direction image may be segmented into any number of slices. In some implementations, the number of slices may be between 8 and 36. However, it should be understood that the number of slices may be less than 8 or greater than 36.

Each of the plurality of slices is representative of at least a portion of the panoramic field of view of the omni-directional image data or the partially panoramic field of view of the omni-directional image data. In some implementations, the plurality of image slices includes a middle image slice (e.g., slice S2), a preceding image slice (e.g., slice S1), and a subsequent image slice (e.g., slice S3), such that a field of view of the middle image slice (e.g., slice S2) is adjacent to or overlaps a preceding field of view of the preceding image slice (e.g., slice S1) and the middle field of view of the middle image slice (e.g., slice S2) is adjacent to or overlaps a subsequent view of the subsequent image slice (e.g., slice S3).

In some implementations, each image slice of the plurality of image slices is representative of an equal portion of the panoramic field of view of the omni-directional image data and the collective fields of view of the plurality of image slices is the same as the panoramic field of view of the omni-directional image data. For example, each of the eight slices captures an eighth of the full panoramic view of the omnidirectional image data and the collective field of view of the eight image slices is the same as the panoramic field of view of the omni-directional image data. In some implementations, the field of view of a first slice of the plurality of views may be greater than a field of view of a second slice of the plurality of slices. In some implementations, the collective fields of view of the plurality of slices may be smaller than the full panoramic field of view. In some implementations, the field of views of neighboring slices may overlap.

The onboard processing array 110 calculates a slice descriptor for each image slice of the plurality of image slices. As used herein, “slice descriptor” refers to a description of the visual features (e.g., color, texture, shape, motion, etc.) of the image data of a particular slice of the omni-directional image data. For example, a slice descriptor d1 is calculated for slice S1, a slice descriptor d2 is calculated for slice S2, a slice descriptor d3 is calculated for slice S3, a slice descriptor d4 is calculated for slice S4, a slice descriptor d5 is calculated for slice S5, a slice descriptor d6 is calculated for slice S6, a slice descriptor d7 is calculated for slice S7, and a slice descriptor d8 is calculated for slice S8.

In some implementations, the slice descriptor may be calculated using an algorithm, such as scale-invariant feature transform (“SIFT”), speeded up robust feature (“SURF”), histogram of oriented gradients (“HOG”), generalized search tree (“GIST”), fast retina keypoint (“FREAK”), and binary robust invariant scalable keypoints (“BRISK”), and the like. However, it should be understood that other algorithms may be used to calculate the slice descriptor. In some implementations, the slice descriptor may include a decimal vector. In some implementations, the slice descriptor may include a binary vector. In other implementations, the slice descriptor may be represented in a format other a binary vector or a decimal vector. Depth information resulting from the application of stereo algorithms may also be used to calculate the slice descriptor.

The onboard processing array 110 generates a current sequence of slice descriptors for the omni-directional image data received. The current sequence of slice descriptors includes the calculated slice descriptor for each image slice of the plurality of image slices. For example, node n1 includes the slice descriptor d1 corresponding to slice S1, node n2 includes the slice descriptor d2 corresponding to slice S2, node n3 includes the slice descriptor d3 corresponding to slice S3, node n8 includes the slice descriptor d8 corresponding to slice S8, etc.

In some implementations, the current sequence of slice descriptors may be structured such that a middle node (e.g., node n2) corresponds to a middle image slice (e.g., slice S2), a preceding node (e.g., node n1) corresponds to a preceding image slice (e.g., slice S1), and a subsequent node (e.g., node n3) corresponds to a subsequent image slice (e.g., slice S3). The preceding node (e.g., node n1) is linked to the middle node (e.g., node n2), and the middle node (e.g., node n2) is linked to the subsequent node (e.g., node n3).

In some implementations, the current sequence of slice descriptors are stored in the storage 113. In some implementations, the storage 113 may include a database of reference sequences of slice descriptors, each of which corresponds to a previously processed omni-directional image encountered by the onboard processing array 110.

In some implementations, the current sequence of slice descriptors may be stored in the storage 113 as a current linked list of slice descriptors. In implementations in which the current sequence of slice descriptors is stored in the storage 113 as a current linked list of slice descriptors, each node of the linked list may be linked to the subsequent node of the linked list (e.g., node n1 is linked to node n2, node n2 is linked to node n3, etc.). In some implementations, the current sequence of slice descriptors may be stored in the storage 113 as a circular linked list of slice descriptors, such that the first node is linked to the second node (e.g., node n1 is linked to node n2), the second node is linked to the third node (e.g., node n2 is linked to node n3), . . . , and the last node is linked back to the first node (e.g., node n8 is linked to node n1). In some implementations, the current sequence of slice descriptors may be stored in the storage 113 as a current doubly linked list of slice descriptors. It should be understood that in other implementations, the current sequence of slice descriptors may be stored in the storage 113 using a data structure other than a linked list, such as an array, and the like.

While the omni-directional image received was not unwarped prior to segmenting the omni-directional image, in other implementations, the omni-directional image may be unwarped prior to segmentation.

Returning to FIG. 4, at block 402, the machine readable instructions stored in the memory 112, when executed by the processor 111, may cause the onboard processing array 110 to access a reference sequence of slice descriptors in the storage 113. For example, in the implementation depicted in FIG. 4, the reference sequence of slice descriptors includes a reference slice descriptor d3 corresponding to a reference node r1, a reference slice descriptor d4 corresponding to a reference node r2, a reference slice descriptor d5 corresponding to a reference node r3, a reference slice descriptor d6 corresponding to a reference node r4, a reference slice descriptor d7 corresponding to a reference node r5, a reference slice descriptor d8 corresponding to a reference node r6, a reference slice descriptor d1 corresponding to a reference node r7, and a reference slice descriptor d2 corresponding to a reference node r8.

Still referring to FIG. 4, at block 404, the machine readable instructions stored in the memory 112, when executed by the processor 111, may cause the onboard processing array 110 to determine whether the current sequence of slice descriptors matches the reference sequence. In some implementations, whether the current sequence of slice descriptors matches the reference sequence of slice descriptors is determined by determining a current order of slice descriptors, determining a reference order of slice descriptors, and comparing the current order of slice descriptors to the reference order of slice descriptors. For example, a current order of slice descriptors in the embodiment depicted in FIG. 4 may be determined as {d1, d2, d3, d4, d5, d6, d7, d8}. A reference order of slice descriptors in the embodiment depicted in FIG. 4 may be determined as {d3, d4, d5, d6, d7, d8, d1, d2}. The current order of slice descriptors {d1, d2, d3, d4, d5, d6, d7, d8} may be compared to the reference order of slice descriptors {d3, d4, d5, d6, d7, d8, d1, d2} in order to determine whether the current order of slice descriptors matches the reference order of slice descriptors.

In some implementations, the current sequence of slice descriptors is a current circular linked list of slice descriptors and the reference sequence of slice descriptors is a reference circular linked list of slice descriptors. In such implementations, the current order of slice descriptors may be determined by traversing the current circular linked list of slice descriptors starting at a current starting node (e.g., the current order of slice descriptors may be determined to be {d1, d2, d3, d4, d5, d6, d7, d8} by traversing the current circular linked list starting from node n1 of the current circular linked list of slice descriptors). The reference order of slice descriptors may be determined by traversing the reference circular linked list of slice descriptors starting at a reference starting node (e.g., the reference order of slice descriptors may also be determined to be {d1, d2, d3, d4, d5, d6, d7, d8} by traversing the reference circular linked list starting from node r7 of the reference circular linked list of slice descriptors). The current sequence of slice descriptors matches the reference sequence of slice descriptors when the current order of slice descriptors is the same as the reference order of slice descriptors. In the embodiment depicted in FIG. 4, the current sequence of slice descriptors may be determined to match the reference sequence of slice descriptors because the reference order of slice descriptors when traversing the reference circular linked list of slice descriptors starting from node r7 is the same as the current order of slice descriptors when traversing the current circular linked list of slice descriptors starting from node n1.

Still referring to FIG. 4, at block 406, the machine readable instructions stored in the memory 112, when executed by the processor 111, may cause the onboard processing array 110 to estimate an orientation or position based on the current sequence of slice descriptors and the reference sequence of slice descriptors. For example, differences between the current sequence of slice descriptors and the reference sequence of slice descriptors may be used to determine a current position or orientation with reference to a known position or orientation associated with the reference sequence of slice descriptors. In some implementations, standard filtering techniques, such as the extended Kalman filter, the particle filter, and the like may be used to determine the current position or orientation based on the comparison between the current sequence of slice descriptors and the reference sequence of slice descriptors.

FIGS. 5A-9 illustrate various possible exemplary implementations of the smart necklace 100. Although FIGS. 5A-9 depict specific locations of components, in other implementations the exact locations and configurations may vary, as more or less components may be added or rearranged depending on specific applications. For example, the onboard processing array 110 may generally be located in the band portion of the smart necklace. Other hidden components, such as the IMU 123, the GPS 124, the sensor 125, the vibration unit 133, the battery 141, and the antenna 142, may be placed along the inside of the smart necklace as needed. The use of specialized sensors may require specific sensor placement and rearrangement of the other components. In addition, although the various implementations described generally take on a U-shape, in other implementations, other shapes, such as an O-shape or Ω-shape, may be used.

Turning to FIGS. 5A-5C, FIGS. 5A-5C depict one implementation of a smart necklace 500 having a “stalk” design. The smart necklace 500 has a left stalk 551 and a right stalk 552, connected by a tube 554, which may be a flexible tube for added comfort. The smart necklace 500 corresponds to the smart necklace 100, and has similar components as the smart necklace 100, which are located within the left stalk 551, the right stalk 552, and the tube 554. Specifically, the tube 554 defines a cavity which houses the unseen components of the smart necklace 500. The left stalk 551 and the right stalk 552 each have a pair of stereo cameras 521, which may be offset by a stereo distance 557. Although in FIG. 5A the stereo distance 557 is the same for both pairs of stereo cameras 521, in other implementations the stereo distance 557 may be different for each pair of stereo cameras 521. Cameras 522 are placed to the sides of the stereo cameras 521 to increase the FOV. Speakers 532 are placed above the left stalk 551 and the right stalk 552. A microphone 531 is placed away from the speakers 532. A button 553 may be used as a power button or on/off switch, or other input. A touch sensor 534, which may be delimited with bump indicators 558, may provide another input for the user. A headphone jack 543 provides a port for attaching, for example, a 3.5 mm headphone set. A data port 563 may provide an additional port for data connections. The headphone jack 543 and the data port 563 may be located on the underside of the smart necklace 500, as seen in FIG. 5C. A texture strip 555, which may be rubber or silicone, lines the tube 554 to provide added cushioning and grip. In addition, the smart necklace 500 may have braille 556 for identification or a label for blind persons.

Referring now to FIGS. 6A-6C, FIGS. 6A-6C depict one implementation of a smart necklace 600 having a “pod” design, which may resemble two peapods connected by a band. The smart necklace 600 has a left pod 651 and a right pod 652, connected by a band 654, which may be a flexible band for added comfort. The smart necklace 600 corresponds to the smart necklace 100, and has similar components as the smart necklace 100, which are located within the left pod 651, the right pod 652, and the band 654. Specifically, the band 654 defines a cavity which houses the unseen components of the smart necklace 500. The left pod 651 and the right pod 652 each have a pair of stereo cameras 621. Cameras 622 are placed to the sides of the stereo cameras 621 to increase the FOV. Each of the stereo cameras 621 and cameras 622 may be placed on its own “pea,” as seen in the three peas in FIGS. 6A and 6C. In other implementations, more or less peas may be used to house additional cameras or sensors. In certain other implementations, one or more of the peas may be removable and modular, to create custom combinations of sensors. For example, additional peas may be screwed on, magnetically attached, or otherwise removably attached, with electrical connector points between adjacent peas for sending power and signals. Speakers 632 are placed above the left pod 651 and the right pod 652. Microphones 631 are placed behind the left pod 651 and the right pod 652 and away from the speakers 632, as seen in FIG. 6B. A headphone jack 643, which may be located on an underside of the smart necklace 600, provides a port for attaching, for example, a 3.5 mm headphone set. A data port 663, which may be a universal serial bus (USB) port or other similar data port, provides an additional data connection to other devices. A touch sensor 634, which may be delimited with bump indicators 658, may provide another input for the user. A button 653 may be used as a power button or on/off switch, or other input. A texture strip 655, which may be rubber or silicone, lines the band 654 to provide added cushioning and grip. In addition, the smart necklace 600 may have braille 656 for identification or a label for blind persons. LEDs 635, placed near the bottom of the left pod 651 and the right pod 652, provide visual indicators, such as status indicators, and may have different colors or flashing patterns to indicate various statuses. An overmold 671, which may be rubber or silicone, lines each of the left pod 651 and the right pod 652, and may be soft and flexible to provide additional grip and cushioning.

FIGS. 7A-7B depict one implementation of a smart necklace 700 having a “paddle” design, which may resemble two paddles connected by a band. The smart necklace 700 has a left paddle 751 and a right paddle 752, connected by a band 754, which may be a flexible band for added comfort. The smart necklace 700 corresponds to the smart necklace 100, and has similar components as the smart necklace 100, which are located within the left paddle 751, the right paddle 752, and the band 754. Specifically, the band 754 defines a cavity which houses the unseen components of the smart necklace 500. The left paddle 751 and the right paddle 752 each have a pair of stereo cameras 721. Cameras 722 are placed to the sides of the stereo cameras 721 to increase the FOV. Speakers 732 are placed above the left paddle 751 and the right paddle 752. Microphones 731 are placed on the sides of the left paddle 751 and the right paddle 752 and towards the inside of the U-shape of the smart necklace 700, as seen in FIG. 7A. A headphone jack 743 provides a port for attaching, for example, a 3.5 mm headphone set. A data port 763 may provide an additional port for data connections. A touch sensor 734, which may be delimited with bump indicators 758, may provide another input for the user. A button 753 may be used as a power button or on/off switch, or other input. An overmold 755, which may be rubber or silicone, may line a portion or a majority of the smart necklace 700 to provide added cushioning and grip. In addition, the smart necklace 700 may have braille 756 for identification or a label for blind persons. LEDs 735, placed near the bottom of the left paddle 751 and the right paddle 752, provide visual indicators, such as status indicators, and may have different colors or flashing patterns to indicate various statuses.

FIGS. 7C-7D illustrate a smart necklace 701, which is similar to the smart necklace 700 with a modified paddle design. A modified paddle 780, which may be either a right paddle or a left paddle or both, has the overmold 755, but narrows at a neck portion 785. The modified paddle 780 is configured to have removable pucks, such as a left puck 781 and a right puck 782 in FIG. 7D. The overmold 755 has a large, oblong puck opening 783 for housing the left puck 781 or the right puck 782. The overmold 755 further includes a side opening 784 to allow the camera 722 to have a side view. The left puck 781 and the right puck 782 may be removed from the overmold 755 through the puck opening 783. In other implementations, the left puck 781 and the right puck 782 may be slid out of an additional hole in the overmold 755, or other similar method as needed. For example, in other implementations the left puck 781 and the right puck 782 may be magnetically attachable, and may further be connected by a retractable tether. The neck portion 785 may contain a connection port for the left puck 781 and the right puck 782 to send signals to and from the smart necklace 700, and to receive power. The left puck 781 and the right puck 782 may further include its own battery, respectively.

In FIG. 7D, the left puck 781 has stereo cameras 721 separated by a stereo distance 757. The right puck 782 has the same stereo distance 757. However, in other implementations the stereo distance 757 may be different. In yet other implementations, the stereo distance 757 may be variable, giving the user the option to remove the left puck 781 and focus on a smaller object, or to get a wider view. For example, the user may wish to have more information about a specific item. By placing the left puck 781 near the item, the stereo cameras 721 may be better able to detect finer details and provide isolated object recognition to provide the user more descriptive information. In certain other implementations, the left puck 781 and/or the right puck 782 may have other sensors, such that the pucks are interchangeable with cameras or sensors as needed in a particular application.

Turning now to FIG. 8, FIG. 8 depicts one implementation of a smart necklace 800 having a “twist” design, which may resemble a twisted band or tube. The smart necklace 800 has a left prong 851 and a right prong 852, connected by a band 854, which may be a flexible band for added comfort. The smart necklace 800 corresponds to the smart necklace 100, and has similar components as the smart necklace 100, which are located within the left prong 851, the right prong 852, and the band 854. The left prong 851 and the right prong 852 each have a pair of stereo cameras 821. Cameras 822 are placed to the sides of the stereo cameras 821 to increase the FOV. An overmold 855, which may be rubber or silicone, may line a portion or a majority of the smart necklace 800 to provide added cushioning and grip. As seen in FIG. 8, the overmold 855 twists around the smart necklace 800, starting from the left prong 851 and twisting around the band 854. An LED 835, placed near the bottom of at least the left prong 851, provides a visual indicator, such as a status indicator, and may have different colors or flashing patterns to indicate various statuses.

Turning now to FIG. 9, FIG. 9 depicts one implementation of a smart necklace 900 having a removable tube 981 with a camera, according to an implementation of the present disclosure. The smart necklace 900 corresponds to the smart necklace 100, and has similar components as the smart necklace 100, which are located within the left portion 951, the right portion 952, and the band 954. The left portion 951 and the right portion 952 may each have a pair of stereo cameras (not shown) similarly as discussed above with respect to FIGS. 5A-8. In one embodiment, the smart necklace 900 may include input devices (e.g., a microphone), output devices (e.g., a speaker and/or haptic feedback), other units, or combinations thereof as described above with respect to embodiments shown in FIGS. 5A-8. Similar to the embodiments described above with respect to FIGS. 7A-7D, a removable tube 981 may be equipped with a camera. For example, the removable tube 981 may be magnetically attachable; attachable with a clip; connected by a retractable tether, wire or cable; or combinations thereof. For example, the user may wish to have more information about a specific item. By placing the camera of the removable tube 981 near the item, the smart necklace 900 may be better able to detect finer details and provide isolated object recognition to provide the user more descriptive information. In other embodiments, removable tubes 981 may be provided on the right portion 952 or both portions 951 and 952, or other portions of the smart necklace 900. In addition to or instead of the removable tube 981, other removable cameras may be provided using various other attaching mechanisms based on practical design and aesthetic concerns in order to allow the user to enhance detection of information.

As used herein, the term “network” includes any cloud, cloud computing system or electronic communications system or method which incorporates hardware and/or software components. Communication among the parties may be accomplished through any suitable communication channels, such as, for example, a telephone network, an extranet, an intranet, Internet, point of interaction device, point of sale device, personal digital assistant (e.g., an Android device, iPhone®, Blackberry®), cellular phone, kiosk, etc.), online communications, satellite communications, off-line communications, wireless communications, transponder communications, local area network (LAN), wide area network (WAN), virtual private network (VPN), networked or linked devices, keyboard, mouse and/or any suitable communication or data input modality. Specific information related to the protocols, standards, and application software utilized in connection with the Internet is generally known to those skilled in the art and, as such, need not be detailed herein.

“Cloud” or “Cloud computing” includes a model for enabling convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction. Cloud computing may include location-independent computing, whereby shared servers provide resources, software, and data to computers and other devices on demand.

Systems, methods and computer program products are provided. References to “various embodiments”, in “some embodiments”, “one embodiment”, “an embodiment”, “an example embodiment”, etc., indicate that the embodiment described may include a particular feature, structure, or characteristic, but every embodiment may not necessarily include the particular feature, structure, or characteristic. Moreover, such phrases are not necessarily referring to the same embodiment. Further, when a particular feature, structure, or characteristic is described in connection with an embodiment, it is submitted that it is within the knowledge of one skilled in the art to affect such feature, structure, or characteristic in connection with other embodiments whether or not explicitly described. After reading the description, it will be apparent to one skilled in the relevant art(s) how to implement the disclosure in alternative embodiments.

The steps of a method or algorithm described in connection with the examples disclosed herein may be embodied directly in hardware, in a software module executed by the processor 111, or in a combination of the two. A software module may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium, such as the storage 113, is coupled to the processor 111 such that the processor 111 can read information from, and write information to, the storage medium. In the alternative, the storage medium may be integral to the processor 111. The processor 111 and the storage medium may reside in an Application Specific Integrated Circuit (ASIC).

The methods/systems may be described herein in terms of functional block components, screen shots, optional selections and various processing steps. It should be appreciated that such functional blocks may be realized by any number of hardware and/or software components configured to perform the specified functions. For example, the methods/systems may employ various integrated circuit components, e.g., memory elements, processing elements, logic elements, look-up tables, and the like, which may carry out a variety of functions under the control of one or more microprocessors or other control devices. Similarly, the software elements of the methods/systems may be implemented with any programming or scripting language such as, VPL, C, C++, C#, Java, JavaScript, VBScript, Macromedia Cold Fusion, COBOL, Microsoft Active Server Pages, assembly, PERL, PHP, awk, Python, Visual Basic, SQL Stored Procedures, PL/SQL, any UNIX shell script, and XML with the various algorithms being implemented with any combination of data structures, objects, processes, routines or other programming elements. Further, it should be noted that the methods/systems may employ any number of conventional techniques for data transmission, signaling, data processing, network control, and the like.

As will be appreciated by one of ordinary skill in the art, the methods/systems may be embodied as a customization of an existing system, an add-on product, upgraded software, a stand-alone system, a distributed system, a method, a data processing system, a device for data processing, and/or a computer program product. Furthermore, the methods/systems may take the form of a computer program product on a non-transitory computer-readable storage medium having computer-readable program code means embodied in the storage medium. Any suitable computer-readable storage medium may be utilized, including hard disks, CD-ROM, optical storage devices, magnetic storage devices, and/or the like.

Exemplary embodiments of the methods/systems have been disclosed in an illustrative style. Accordingly, the terminology employed throughout should be read in a non-limiting manner. Although minor modifications to the teachings herein will occur to those well versed in the art, it shall be understood that what is intended to be circumscribed within the scope of the patent warranted hereon are all such embodiments that reasonably fall within the scope of the advancement to the art hereby contributed, and that that scope shall not be restricted, except in light of the appended claims and their equivalents.

Claims

1. A wearable neck device for providing environmental awareness to a user, comprising: a flexible tube for being worn around a neck of the user, defining a cavity, having a U-shape, and having a center portion configured to rest on a back of a neck of the user, a left portion and a right portion;a first stereo pair of cameras positioned on the left portion of the flexible tube;a second stereo pair of cameras positioned on the right portion of the flexible tube;an inertial measurement unit (IMU) positioned within the cavity and configured to detect inertial measurement data;a vibration unit positioned within the cavity and configured to provide haptic feedback to the user;a speaker positioned along the flexible tube and configured to provide audio feedback to the user; anda processor positioned within the cavity and configured to receive video frames from the first stereo pair of cameras and the second stereo pair of cameras, receive the detected inertial measurement data from the IMU, provide object recognition of items in the video frames, identify points of interest to the user based on the object recognition, determine navigation instructions based on the inertial measurement data and the received video frames, and control the vibration unit to provide haptic feedback and the speaker to provide audio feedback to the user based on the points of interest and the navigation instructions.

2. The wearable neck device of claim 1 further comprising a storage device configured to store a map data.

3. The wearable neck device of claim 1 further comprising an antenna configured for wireless communication.

4. The wearable neck device of claim 1 further comprising at least one of a radio frequency identification (RFID) reader for reading RFID tags or a near field communication (NFC) unit for reading NFC tags.

5. The wearable neck device of claim 1 further comprising a global position system (GPS) unit positioned in the cavity and configured to detect location data and wherein the processor is further configured to determine the navigation instructions based on the location data.

6. The wearable neck device of claim 1 further comprising a microphone for receiving voice commands.

7. The wearable neck device of claim 1 further comprising a battery capable of being wirelessly charged.

8. The wearable neck device of claim 1 further comprising an input device for adjusting parameters for haptic and audio feedback.

9. The wearable neck device of claim 1 further comprising at least one of a fifth camera positioned on the left portion of the flexible tube or a sixth camera positioned on the right portion of the flexible tube, the at least one of the fifth camera or the sixth camera being configured to increase a field of view for the wearable neck device.

10. The wearable neck device of claim 1 wherein the first stereo pair of cameras is removable from the left portion of the flexible tube and the second stereo pair of cameras is removable from the right portion of the flexible tube.

11. The wearable neck device of claim 1 wherein the first stereo pair of cameras is coupled to the flexible tube using at least one of a retractable tether, a string or a cable.

12. The wearable neck device of claim 10 wherein at least one of the first stereo pair of cameras or the second stereo pair of cameras are capable of isolated object recognition when placed near an item.

13. The wearable neck device of claim 1 further comprising a removable tube having a fifth camera, the removable tube being coupled to at least one of the left portion or the right portion using at least one of a magnetic attachment; a clip; a retractable tether, a string or a cable, wherein the fifth camera is capable of isolated object recognition when placed near an item.

14. A wearable neck device for providing environmental awareness to a user, comprising: a band defining a cavity and having a center portion, a left portion and a right portion;a first stereo pair of cameras positioned on the left portion of the band;a first camera positioned to a side of the first stereo pair of cameras;a second stereo pair of cameras positioned on the right portion of the band;a second camera positioned to a side of the second stereo pair of cameras;an inertial measurement unit (IMU) coupled to the band and configured to detect inertial measurement data;a vibration unit positioned within the cavity and configured to provide haptic feedback to the user;a speaker positioned along the band and configured to provide audio feedback to the user; anda processor positioned within the cavity and configured to receive video frames from the first stereo pair of cameras, the first camera, the second stereo pair of cameras and the second camera, receive the inertial measurement data from the IMU, provide object recognition of items in the video frames, identify points of interest to the user based on the object recognition, determine navigation instructions based on the inertial measurement data and the video frames, and control the vibration unit to provide haptic feedback and the speaker to control the speaker to provide audio feedback to the user based on the items of interest and the navigation instructions.

15. The wearable neck device of claim 14 further comprising an overmold covering at least a portion of the band to provide cushioning and grip.

16. The wearable neck device of claim 14, further comprising: a storage device configured to store a map data, anda global position system (GPS) unit coupled to the band and configured to detect location data,wherein the processor is further configured to determine the navigation instructions based on the location data.

17. The wearable neck device of claim 14 further comprising an antenna configured for wireless communication.

18. The wearable neck device of claim 14 wherein the first stereo pair of cameras is removable from the left portion of the band and the second stereo pair of cameras is removable from the right portion of the band to provide isolated object recognition when the first stereo pair of cameras or the second stereo pair of cameras is placed near an item.