Generating and reviewing motion metadata

Abstract

Aspects of the instant disclosure relate to methods for generating motion metadata for a newly captured video feed. In some aspects, methods of the subject technology can include steps for recording a video feed using the video capture system, partitioning the image frames into a plurality of pixel blocks, and processing the image frames to detect one or more motion events. In some aspects, the method may further include steps for generating motion metadata describing each of the one or more motion events. Systems and computer-readable media are also provided.

Description

TECHNICAL FIELD

The disclosed technology relates to systems and methods for facilitating the retroactive review of video data and in particular, for performing real-time processing on a recorded video feed and storing motion metadata for the subsequent search of motion events.

BACKGROUND

The proliferation of video cameras, such as those deployed in security surveillance setups, has drastically increased the amount of collected video data. Because human review of camera footage is tiresome and expensive, some processing techniques have been developed that enable the pre-selection of areas-of-interest within the feed for which processing is to be performed. However, such processing methods require that the user identify regions of interest before recording, a constraint that is often impractical for security deployments in which a location of events-of-interest within the frame is often hard to predict.

BRIEF DESCRIPTION OF THE DRAWINGS

The embodiments herein may be better understood by referring to the following description in conjunction with the accompanying drawings in which like reference numerals indicate similar elements. It is understood that these drawings depict only example embodiments of the disclosure and are not to be considered limiting in scope. The principles herein are described and explained with additional specificity and detail through the use of the accompanying drawings in which:

FIG. 1 illustrates an example environment in which some aspects of the technology can be implemented.

FIG. 2A illustrates an example of an image frame that is parsed into coarse-cells for facilitating motion detection, according to some aspects of the technology.

FIG. 2B illustrates an example image frame depicted in FIG. 2A, that is parsed into macro-blocks for facilitating motion detection, according to some aspects of the technology.

FIG. 2C illustrates an example use of macro-blocks to indicate areas of motion activity, as well as a corresponding motion vector, according to some aspects of the technology.

FIG. 3A illustrates steps of an example process for associating motion metadata with recorded video information.

FIG. 3B illustrates of an example process for receiving a motion event search query, and performing a search through motion metadata to identify one or more motion events.

FIG. 4 illustrates an example processor-based device on which aspects of the technology can be implemented.

OVERVIEW

Aspects of the instant disclosure relate to methods for generating motion metadata for a newly captured video feed. In some aspects, a process of the technology can include steps for recording a video feed (e.g., storing video data) using a video capture system, partitioning image frames of the video feed into a multiple pixel blocks, and processing the image frames to detect one or more motion events. In some aspects, the process can further include steps for generating motion metadata describing each of the motion events.

DESCRIPTION

Aspects of the disclosed technology address some of the limitations of searching for motion events in legacy video content by providing systems/methods to facilitate the real-time processing of motion events within a recorded video feed. In some aspects, motion events are associated with frame and/or frame-location before storage is completed. As such, the disclosed technology facilitates the quick and easy searching of motion events in stored video data, including the selection and searching of motion events within a frame-subset that defines and area within the video feed that can be selected for the search of motion events.

FIG. 1 illustrates an example environment 100 in which some aspects of the technology can be implemented. Environment 100 includes an image capture device 102 configured for recording video information (e.g., a video feed) as a series of image frames 104. Environment 100 further includes an image processing module 106, metadata/video database 108, and a video server 110. Additionally, environment 100 includes a user 114A, for example, that is associated with user device 114B, and communicatively coupled to video server 110, via network 105.

Environment 100 conceptually illustrates various hardware and/or software modules that can be used to implement aspects of the technology; however, it is understood that additional hardware and/or software modules can be used without departing from the scope of the technology. For example, network 105 can represent a private network such as a local area network (LAN), a wide area network (WAN), or a network of public and/or private networks, such as the Internet. Additionally, it is understood that the various software modules (e.g., image processing module 106, metadata/video database 108, and/or metadata search module 112) can be implemented on one or more hardware devices, such as an image capture device 102, or one or more computers/servers, such as server 110.

In operation, image capture device 102 is configured to record and store video data in a series of image (or video) frames, e.g. frames 104. Although image capture device 102 can include any video recording device capable of capturing or recording and storing digital video information, by way of example, image capture device 102 can be a video surveillance camera, or the like. Additionally, storage frames 104 can be performed locally (e.g., on a memory device visually coupled with image capture device 102), or made from remotely, such as via storage on one or more remote servers communicatively coupled with image capture device 102 via network 105.

Image frames 104 can be processed and stored as video capture is performed. For example, image capture device 102 can be configured to provide a real-time feed of captured image frames 104 to an image processing module configured to detect motion events. As discussed in further detail below, motion events can be detected by calculating a sum of absolute differences between frames on a frame-by-frame basis. Detected motion events can then be indexed based on various indicators, such as, a time stamp, duration, and/or an intensity associated with the frame. Subsequently, data describing the detected motion events (i.e. “motion metadata”) is stored in a database, such as metadata/video database 108.

It is understood that image processing module 106 and/or metadata/video database 108 can be implemented on a common hardware platform, such image capture device 102, and/or a computer/server such as video server 110. Similarly, any of the described software modules environment 100 may be operated using one or more servers that are communicatively coupled via network 105, for example, as various software processes instantiated in a computing cluster (not illustrated).

In practice, metadata/video database 108 is accessible by a metadata search module 112 that can be used to facilitate queries relating to motion metadata for a particular set of image frames 104.

Once motion metadata has been stored, a user (e.g., user 114A) can provide a motion search query (e.g., via user device 114B) to video server 110 over a computer network, such as network 105. The search query can then be received by metadata search module 112 and used to search metadata/video database 108. As discussed in further detail below with respect to FIG. 3B, the motion search query can include boundary conditions indicating a particular area or region within the video frames in which motion events are to be searched. By permitting users to specify a particular “region of interest” for their motion search query, legacy video data can be efficiently searched only for events occurring at indicated locations of interest, increasing the speed and efficiency of the search, as well as reducing processing overhead.

By way of example, a security surveillance system can capture video footage that includes the theft of an item of interest located within a particular region within the image frame. Thus, a user may wish to only search for motion events that occur within the particular region of interest (e.g., at a location where the item was located). In such instances, the user may provide a search query in the form of a selection of the image frame area for which motion events are to be searched, for example, by indicating the area with a cursor or other input device. The search query can then be used to search motion metadata stored in a time-series database, such as that hosted by video server 110.

As discussed in further detail below, the time-series database of motion metadata can be comprised of a series of motion vectors that described various attributes of frames within the video feed. Although the motion vectors can describe essentially any attribute of an associated image frame, in some approaches, motion vectors contain information regarding the location of motion events as placed in a coarse-cell and/or macro-block (e.g., “fine cell”) grid. As discussed in further detail with respect to FIG. 2A, coarse/fine-cell grids can be superimposed over an image area, effectively providing a bit mask that permits irrelevant information to be efficiently disregarded.

Specifically, FIG. 2A illustrates an example of an image frame 201 that is parsed into coarse-cells 203. In the example of FIG. 2A, coarse-cells 203 are arranged in a 9×10 grid superimposed over image frame 201. However, it is understood that other coarse-cell overlay dimensions can be used, without departing from the scope of the technology. As discussed in further detail below, motion detection can be performed using subdivided units of coarse-cells 203 (i.e., macro-blocks), which represent the smallest collection of adjacent pixels for which motion detection events are registered. As discussed in greater detail below, motion events associated with a particular macro-block can be identified by calculating absolute differences (e.g., in pixel color or intensity) as between common macro-blocks in adjacent frames, i.e., on a frame-to-frame basis.

FIG. 2B illustrates an example of image frame 201 depicted in FIG. 2A, wherein each of coarse-cells 203 are further parsed into macro-blocks 205. Macro-blocks 205 can be used to distinguish regions within a particular coarse-cell in which motion is detected. In the example of FIG. 2B, each coarse-cell 203 contains twenty-four total macro-blocks 205; however, it is understood that any other number of macro-blocks may be used to subdivide a coarse-cell, without departing from the technology.

Calculations of motion events can be determined based on changes in pixel values for a particular macro-block across multiple image frames. In the example of FIG. 2B, macro-blocks 205 cover an area of image frame 201 (e.g., coarse-cell A1) that does not contain any motion events. In contrast, FIG. 2C illustrates an example of a coarse-cell (D7) in which a motion event is detected.

Specifically, FIG. 2C illustrates an example of macro-blocks 205 used to indicate locations of motion activity, for example, in coarse-cell D7, as well as a corresponding motion vector 207. In some aspects, motion vectors, such as motion vector 207, can be used to provide a compressed data format for storing motion event data.

As indicated in the example of FIG. 2C, motion events within coarse-cell D7 are identified at macro-blocks 13, 10, 14, 18, 7, 11, 15, and 19. Motion events for each macro-block are described by motion vector 207 that provides an indication of the row/column of the associated coarse-cell, as well as the associated macro-blocks in which motion has been detected. Motion vectors, such as motion vector 207, can be associated with a respective image frame that is sorted as a times-series, for example, in a database such as metadata/video database 108, discussed above.

In some aspects, the motion vector depicted by motion vector 207 is advantageous in that each macro-block can be represented as a single 32-bit integer, wherein the first four bits indicate the macro-block row, and the next four bits indicate the macro-block column. In this example, the remaining 24 bits act as a bit-mask used indicate the presence of motion within each of the macro-block cells. Therefore, in the example of FIG. 2C, the entire image can be stored as up to 90 of these 32-bit integers (e.g., 360 bytes).

In some aspects, the motion vector format also permits sparse data to be vastly compressed. For example, macro-blocks that contain no motion events are not saved, and a motion event that activates only a single macro-block cell consumes only 4 bytes of data.

FIG. 3A illustrates steps of an example process 300 for associating motion metadata with recorded video information. Process 300 begins with step 302, in which a video feed is captured using a video capture system, such as image capture device 102 discussed above with respect to FIG. 1. The captured video feed includes multiple image frames each of a predefined area (e.g., a predefined dimension).

In step 304, the video feed is partitioned into a plurality of pixel blocks, such as coarse-cells, and macro-blocks (fine cells), as illustrated above with respect to FIG. 2A-C. It is understood that video feed partitioning can be performed at the image capture device (e.g. a security camera), and/or at one or more computing devices, such as one or more computers/servers and/or a computing cluster.

At step 306, the image frames are processed to detect one or more motion events. Each detected motion event is associated with at least one pixel block (e.g., a macro-block). As discussed above, determinations of motion events can be based on a calculation of a sum of absolute differences (SAD) between individual macro-blocks (also “frame subtraction”). In other approaches, different types of background subtraction calculations can be implemented; however, it is understood that other motion detection methods may be used, without departing from the scope of the technology.

In step 308, motion metadata is generated based on the one or more motion events detected at step 306. As discussed above, motion metadata may be generated as a motion vector, (e.g., motion vector 207), that is used to describe various attributes associated with the motion event. Such attributes can include indications of time, coarse-block location, macro-block location, and/or intensity, etc.

Next, in step 310, the motion metadata is associated with one or more image frames and/or one or more specific pixel blocks within a respectively associated frame. As discussed in further detail below, the association of motion metadata with image frames can facilitate the convenient searching of metadata events in legacy content, for example, within a region of the image frame specified after the video feed has already been recorded.

FIG. 3B illustrates of an example process 301 for performing a search through motion metadata to identify one or more motion events. Further to the example of step 310, with respect to process 300, the searching of motion metadata in process 301 begins with step 312 in which a motion event search query is received, for example, from a user or system administrator. The motion event search query can be received by a metadata search module (e.g., metadata search module 112) configured to access the metadata/video database where motion metadata is stored, such as metadata/video database 108.

Although the format of the motion event search query can depend on the desired implementation, in some aspects the motion event search query can contain information that defines at least a portion of the predefined pixel area for which the desired search is to be performed. For example, the motion event search query can take the form of a bounded area defining at least a portion of the image frame across which all of the video content is to be searched.

In some aspects, the motion event search query can contain indications of boundaries within the video feed. For example, the motion event search query may specify an area of the image frame, in conjunction with start and stop times (corresponding with different image frames or frame numbers in the feed), for which the search is to be performed. Other search parameters can include, but are not limited to, searches based on event length and/or intensity (e.g., to filter out sudden increases/decreases in ambient lighting). It is understood that other user-specified criteria may be included in the motion event search query, without departing from the scope of the technology.

Subsequently, at step 314, the motion metadata is searched to identify one or more image frames that include motion events of interest. In some aspects, the search query may return more specific information, such as, identifications of a specific pixel block (e.g., macro-block and/or micro-block), that include motion events of interest.

Further to the example provided above with respect to FIGS. 2A-2C, if a user specifies an area of the image frame that includes macro-block D7, the search could be performed across all macro-blocks corresponding with the user-specified area, across all image frames. In this example, because motion events have been identified to occur in those blocks, for example, at least in block D7 (as specified by motion vector 207), the associated frames and/or blocks would be returned to the user for further review.

EXAMPLE DEVICE

FIG. 4 illustrates an example processor-based device 400 that can be used to implement various aspects of the technology. For example, processor-based device 400 may be used to implement an image capture device (e.g., image capture device 102), or a video server, such as video server 110 discussed above with respect to FIG. 1. It is further understood that processor-based device 400 may be used in conjunction with one or more other processor-based devices, for example, as part of a computer network or computing cluster.

Processor-based device 400 includes a master central processing unit (CPU) 462, interfaces 468, and a bus 415 (e.g., a PCI bus). When acting under the control of appropriate software and/or firmware, the CPU 462 can be configured for executing operations necessary to process video content to determine motion events within one or more image frames, as described above.

CPU 462 preferably accomplishes all these functions under the control of software including an operating system and any appropriate applications software. CPU 462 can include one or more processors 463 such as a processor from the Motorola family of microprocessors or the MIPS family of microprocessors. In an alternative embodiment, processor 463 is specially designed hardware for controlling the operations of processor-based device 400. In a specific embodiment, a memory 461 (such as non-volatile RAM and/or ROM) also forms part of CPU 462. However, there are many different ways in which memory could be coupled to the system.

Interfaces 468 can be provided as interface cards (sometimes referred to as “line cards”). Generally, they control the sending and receiving of data packets over the network and sometimes support other peripherals used with the router 410. Among the interfaces that may be provided are Ethernet interfaces, frame relay interfaces, cable interfaces, DSL interfaces, token ring interfaces, and the like. In addition, various very high-speed interfaces may be provided such as fast token ring interfaces, wireless interfaces, Ethernet interfaces, Gigabit Ethernet interfaces, ATM interfaces, HSSI interfaces, POS interfaces, FDDI interfaces and the like. Generally, these interfaces may include ports appropriate for communication with the appropriate media. In some cases, they may also include an independent processor and, in some instances, volatile RAM. The independent processors may control such communications intensive tasks as packet switching, media control and management. By providing separate processors for the communications intensive tasks, these interfaces allow the master microprocessor 462 to efficiently perform routing computations, network diagnostics, security functions, etc.

Although the system shown in FIG. 4 is one specific network device of the present invention, it is by no means the only device architecture on which the present invention can be implemented. For example, an architecture having a single processor that handles communications as well as routing computations, etc. is often used. Further, other types of interfaces and media could also be used with the router.

Regardless of the network device's configuration, it may employ one or more memories or memory modules (including memory 461) configured to store program instructions for the general-purpose network operations and mechanisms for roaming, route optimization and routing functions described herein. The program instructions may control the operation of an operating system and/or one or more applications, for example. The memory or memories may also be configured to store tables such as mobility binding, registration, and association tables, etc.

For clarity of explanation, in some instances the present technology may be presented as including individual functional blocks including functional blocks comprising devices, device components, steps or routines in a method embodied in software, or combinations of hardware and software.

In some embodiments the computer-readable storage devices, mediums, and memories can include a cable or wireless signal containing a bit stream and the like. However, when mentioned, non-transitory computer-readable storage media expressly exclude media such as energy, carrier signals, electromagnetic waves, and signals per se.

Methods according to the above-described examples can be implemented using computer-executable instructions that are stored or otherwise available from computer readable media. Such instructions can comprise, for example, instructions and data which cause or otherwise configure a general purpose computer, special purpose computer, or special purpose processing device to perform a certain function or group of functions. Portions of computer resources used can be accessible over a network. The computer executable instructions may be, for example, binaries, intermediate format instructions such as assembly language, firmware, or source code. Examples of computer-readable media that may be used to store instructions, information used, and/or information created during methods according to described examples include magnetic or optical disks, flash memory, USB devices provided with non-volatile memory, networked storage devices, and so on.

Devices implementing methods according to these disclosures can comprise hardware, firmware and/or software, and can take any of a variety of form factors. Typical examples of such form factors include laptops, smart phones, small form factor personal computers, personal digital assistants, rackmount devices, standalone devices, and so on. Functionality described herein also can be embodied in peripherals or add-in cards. Such functionality can also be implemented on a circuit board among different chips or different processes executing in a single device, by way of further example.

The instructions, media for conveying such instructions, computing resources for executing them, and other structures for supporting such computing resources are means for providing the functions described in these disclosures.

Although a variety of examples and other information was used to explain aspects within the scope of the appended claims, no limitation of the claims should be implied based on particular features or arrangements in such examples, as one of ordinary skill would be able to use these examples to derive a wide variety of implementations. Further and although some subject matter may have been described in language specific to examples of structural features and/or method steps, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to these described features or acts. For example, such functionality can be distributed differently or performed in components other than those identified herein. Rather, the described features and steps are disclosed as examples of components of systems and methods within the scope of the appended claims. Moreover, claim language reciting “at least one of” a set indicates that one member of the set or multiple members of the set satisfy the claim.

Claims

1. A video capture system, comprising: one or more processors; anda non-transitory memory coupled to the processors, the memory storing instructions configured to cause the processors to perform operations comprising: recording a video feed using the video capture system, the video feed comprising a plurality of image frames of a predefined pixel area;partitioning the image frames into a plurality of pixel blocks, wherein each of the pixel blocks comprises a subset of the predefined pixel area of the corresponding image frame;processing each of the plurality blocks of the image frames to detect one or more motion events that occurred across one or more image fames in the corresponding pixel block; andgenerating motion metadata describing each of the one or more motion events;wherein the processing detects motion events within the image frames without being previously directed to specific areas of the image frame.

2. The video capture system of claim 1, further comprising: associating the motion metadata with one or more pixel blocks, wherein the motion metadata associated with each of the pixel blocks corresponds with motion events that occurred across one or more image fames in the corresponding pixel block.

3. The video capture system of claim 1, further comprising: receiving a motion event search query, wherein the search query specifies at least a portion of the predefined pixel area for which a search is to be performed for motion events in the recorded video feed; andsearching the motion metadata to identify one or more image frame and pixel block pairs that include motion events corresponding with the search query.

4. The video capture system of claim 3, wherein the search query specifies a start time and an end time for which motion metadata of the recorded video feed is to be searched.

5. The video capture system of claim 1, wherein processing the image frames to detect one or more motion events further comprises: quantifying motion events for one or more of the pixel blocks; anddetermining if the quantity of motion in each of the one or more pixel blocks exceeds a predetermined threshold.

6. The video capture system of claim 5, wherein the motion events are quantified based on an absolute difference of pixels between two or more image frames of the video feed.

7. The video capture system of claim 1, wherein the motion metadata comprises one or more of: duration information, time stamp information, or spatial data.

8. A method of generating motion metadata, the method comprising: recording a video feed using the video capture system, the video feed comprising a plurality of image frames of a predefined pixel area;partitioning the image frames into a plurality of pixel blocks, wherein each of the pixel blocks comprises a subset of the predefined pixel area of the corresponding image frame;processing each of the plurality blocks of the image frames to detect one or more motion events that occurred across one or more image fames in the corresponding pixel block; andgenerating motion metadata describing each of the one or more motion events;wherein the processing detects motion events within the image frames without being previously directed to specific areas of the image frame.

9. The method of claim 8, further comprising: associating the motion metadata with one or more pixel blocks, wherein the motion metadata associated with each of the pixel blocks corresponds with motion events that occurred across one or more image fames in the corresponding pixel block.

10. The method of claim 8, further comprising: receiving a motion event search query, wherein the search query specifies at least a portion of the predefined pixel area for which a search is to be performed for motion events in the recorded video feed; andsearching the motion metadata to identify one or more image frame and pixel block pairs that include motion events corresponding with the search query.

11. The method of claim 10, wherein the search query specifies a start time and an end time for which motion metadata of the recorded video feed is to be searched.

12. The method of claim 8, wherein processing the image frames to detect one or more motion events further comprises: quantifying motion events for one or more of the pixel blocks; anddetermining if the quantity of motion in each of the one or more pixel blocks exceeds a predetermined threshold.

13. The method of claim 12, wherein the motion events are quantified based on an absolute difference of pixels between two or more image frames of the video feed.

14. The method of claim 8, wherein the motion metadata comprises one or more of: duration information, time stamp information, or spatial data.

15. A non-transitory computer-readable storage medium comprising instructions stored therein, which when executed by one or more processors, cause the processors to perform operations comprising: recording a video feed using the video capture system, the video feed comprising a plurality of image frames of a predefined pixel area;partitioning the image frames into a plurality of pixel blocks, wherein each of the pixel blocks comprises a subset of the predefined pixel area of the corresponding image frame;processing each of the plurality blocks of the image frames to detect one or more motion events that occurred across one or more image fames in the corresponding pixel block; andgenerating motion metadata describing each of the one or more motion events;wherein the processing detects motion events within the image frames without being previously directed to specific areas of the image frame.

16. The non-transitory computer-readable storage medium of claim 15, wherein the instructions are further configured to cause the processors to perform operations comprising: associating the motion metadata with one or more pixel blocks, wherein the motion metadata associated with each of the pixel blocks corresponds with motion events that occurred across one or more image fames in the corresponding pixel block.

17. The non-transitory computer-readable storage medium of claim 15, wherein the instructions are further configured to cause the processors to perform operations comprising: receiving a motion event search query, wherein the search query specifies at least a portion of the predefined pixel area for which a search is to be performed for motion events in the recorded video feed; andsearching the motion metadata to identify one or more image frame and pixel block pairs that include motion events corresponding with the search query.

18. The non-transitory computer-readable storage medium of claim 17, wherein the search query specifies a start time and an end time for which motion metadata of the recorded video feed is to be searched.

19. The non-transitory computer-readable storage medium of claim 15, wherein the instructions are further configured to cause the processors to perform operations comprising: quantifying motion events for one or more of the pixel blocks; anddetermining if the quantity of motion in each of the one or more pixel blocks exceeds a predetermined threshold.

20. The non-transitory computer-readable storage medium of claim 15, wherein the motion events are quantified based on an absolute difference of pixels between two or more image frames of the video feed.