Patent Grant
10152778
The present application is related to PCT Application No. PCT/CN2015/089441, filed on Sep. 11, 2015, and titled “SCALABLE REAL-TIME FACE BEAUTIFICATION OF VIDEO IMAGES”.
When taking still images, users might desire for some levels of control over their appearance. Such desire has lead to face beautification apps being one of the most popular apps for smart phones.
Video sharing and conferencing has been increasingly used with the pervasive usage of smart phone. However, many of the existing apps for smart phones are designed for off-line image processing or might not work with out limited features in a video mode.
The material described herein is illustrated by way of example and not by way of limitation in the accompanying figures. For simplicity and clarity of illustration, elements illustrated in the figures are not necessarily drawn to scale. For example, the dimensions of some elements may be exaggerated relative to other elements for clarity. Further, where considered appropriate, reference labels have been repeated among the figures to indicate corresponding or analogous elements. In the figures:
While the following description sets forth various implementations that may be manifested in architectures such system-on-a-chip (SoC) architectures for example, implementation of the techniques and/or arrangements described herein are not restricted to particular architectures and/or computing systems and may be implemented by any architecture and/or computing system for similar purposes. For instance, various architectures employing, for example, multiple integrated circuit (IC) chips and/or packages, and/or various computing devices and/or consumer electronic (CE) devices such as set top boxes, smart phones, etc., may implement the techniques and/or arrangements described herein. Further, while the following description may set forth numerous specific details such as logic implementations, types and interrelationships of system components, logic partitioning/integration choices, etc., claimed subject matter may be practiced without such specific details. In other instances, some material such as, for example, control structures and full software instruction sequences, may not be shown in detail in order not to obscure the material disclosed herein.
The material disclosed herein may be implemented in hardware, firmware, software, or any combination thereof. The material disclosed herein may also be implemented as instructions stored on a machine-readable medium, which may be read and executed by one or more processors. A machine-readable medium may include any medium and/or mechanism for storing or transmitting information in a form readable by a machine (e.g., a computing device). For example, a machine-readable medium may include read only memory (ROM); random access memory (RAM); magnetic disk storage media; optical storage media; flash memory devices; electrical, optical, acoustical or other forms of propagated signals (e.g., carrier waves, infrared signals, digital signals, etc.), and others.
References in the specification to “one implementation”, “an implementation”, “an example implementation”, etc., indicate that the implementation described may include a particular feature, structure, or characteristic, but every implementation may not necessarily include the particular feature, structure, or characteristic. Moreover, such phrases are not necessarily referring to the same implementation. Further, when a particular feature, structure, or characteristic is described in connection with an implementation, it is submitted that it is within the knowledge of one skilled in the art to effect such feature, structure, or characteristic in connection with other implementations whether or not explicitly described herein.
Systems, apparatus, articles, and methods are described below including operations for real-time face beautification features for video images.
As described above, video sharing and conferencing has been increasingly used with the pervasive usage of smart phone. However, many of the existing apps for smart phones are designed for off-line image processing or might not work with out limited features in a video mode.
However, as will be described in greater detail below, some of the implementations described herein may present a new set of face beautification features including cute nose, happy/sad face, eye wrinkles removal, eye bags removal, and/or dark eye circles removal. The fundamental techniques enabling face beautification features utilized herein are morphological operation and image enhancement where both are applied based on analytic information on the face area. The information utilized from face analysis includes face shape, facial landmark positions and skin tone likelihood/score.
Some of the implementations described herein may utilize a complete face beautification (FB) pipe. The analytic information utilized in such a FB pipe may include face shape, facial landmark points and a skin tone score. The pipe may combine information of face shape, facial landmark points and skin tone score and may utilizes a GPU sampler engine fully.
The face beautification (FB) pipe may be an intelligent FB solution where customized processing is embedded. That is, the features with corresponding levels of processing to be applied to users differentiated by gender/age/racial under different environments are selected automatically and visually suitable/pleasant augmented results are obtained.
In video pre-processing, the input video 101 (captured by camera or video clip) is firstly sent to the GPU video augmentation pipe 100 for skin tone detection (see VEBox-STD 102 and box filter 104) as well as facial shape/feature extraction (face detection 106 and landmark detection 108). The skin tone detection may be done on GPU fixed function (video enhancement box 102). At the same time, the facial shape/feature extraction kernels 106/108 (GPGPU kernel) may be dispatched to execution unit on GPU.
In the application of skin tone based filters, skin tone based filters may include face brightening 110, face whitening 112, skin tone enhancement 114, skin foundation 116, skin smoothing 118, the like, and/or combinations thereof (Note: the feature set may be expanded and/or the order of this feature set might be changed). The processed skin tone map from skin tone detection and box filter modules 102/104 will be consumed by these filters. Algorithm flow charts of skin foundation filter 116 and skin smoothing filter 118 are illustrated below.
In the application of facial landmark based filters, the landmark based filters include red lip filter 120, big eyes filter 122, slim face filter 124, cute nose filter 126, happy/sad face filter 128, eye wrinkle remover filter 130, eye bags remover filter 132, dark eye circles remover filter 134, the like, and/or combinations thereof. Algorithm flow charts of the algorithm flow charts of red lip filter 120, big eyes filter 122, slim face filter 124 are illustrated below.
The face recognition/profile detection filter 140 may be optional. If face recognition/profile detection filter 140 is turned on, face recognition/profile detection will be used to customize the feature set. For example, when gender detection is on, red lip filter 120 may be turned off for male. Further, when face recognition/profile detection filter 140 is on, a user may be able to customize the setting (filter on/off, filter strength, etc.) and save the configuration for future use. That is, the features with corresponding levels of processing to be applied to users may be differentiated by gender/age/racial under different environments are selected automatically and visually suitable/pleasant augmented results are obtained.
The processed picture 141 may be shown on a display and/or encoded and transmitted.
In operation, the video augmentation pipe 100 may leverage existing fixed-function STDE hardware (e.g., Skin Tone Detection and Enhancement 102 module) on a GPU, which may be low-power and high-performance to generate the skin-tone likelihood. For further information of such STDE hardware (e.g., Skin Tone Detection and Enhancement module), e.g., please refer to U.S. Pat. No. 8,493,402, filed Jan. 6, 2011, and titled “System, method and computer program product for color processing of point-of-interest color”.
The generation of facial landmark points, on the other hand, may leverages both a face detector, e.g., please refer to PCT Application No. PCT/CN2014/075165, filed Apr. 11, 2014, entitled “OBJECT DETECTION USING DIRECTIONAL FILTERING”, and fixed-function hardware (e.g., Convolution Filter), e.g., please refer to PCT Application No. PCT/CN2012/086840, filed Dec. 18, 2012, entitled “HARDWARE CONVOLUTION PRE-FILTER TO ACCELERATE OBJECT DETECTION”, for the face detection 106 task, which may be performed prior to facial landmark 108 detection. The face detection 106 approach taken here can achieve fast-compute and low-power target while maintaining excellent detection accuracy.
Furthermore, a light-weight compute of facial landmark validation step may be embedded in the facial landmark 108 generation block to intelligently switch between detection and tracking mode. This automatic switch between tracking and detection mode may help reduce the compute while maintaining good accuracy in terms of locating facial landmark.
Some of the implementations described herein may combine GPU based face detection, skin tone detection, facial landmark detection and fully utilizes GPU hardware (e.g., Sampler engine, EU, Fixed Function, etc.) to build one power efficient real-time face beautification pipe on HD-resolution video.
For example, in experiments run at HD (1280×720) on an Intel GPU @400 MHz, a face beautification video augmentation pipe 100, as described herein, achieve real-time (e.g., 30 fps) on HD video with very low CPU utilization and can use multiple face beautification filters simultaneously.
Additional details regarding the functioning of face beautification (FB) video augmentation pipe 100 are illustrated below with regard to
The Skin-Tone Detection (STD) 102 utilizes Video Enhancement Box (VEBox) Skin-Tone Detection and Enhancement (STDE) in a graphics processor to perform skin tone color detection for an input YUV data. The output of the STD contains is skin-tone likelihood score represented in 5-bit for each pixel within an input frame.
The input of the Box-Filter 104 is the skin-tone score result from Skin-Tone Detection (STD) 102 (e.g., VEBox). Box-Filter 104 performs averaging operation on the input skin-tone score to produce a smooth version of skin-tone likelihood score.
The Face Detection 106 takes YUV input and applies a pre-trained model, which only operates on Y-channel information to identify the appearance of human faces within an input frame. Face Detection 106 returns the location and size of each detected face. In some implementations, face detection 106 may be implemented as a combination of hardware and software solutions.
The Facial Landmark Detection/Tracking 108 takes YUV input and information of detected faces from the Face Detection 106. Facial Landmark Detection/Tracking 108 applies a pre-trained model on the rectangle area of each detected face to detect/track the locations of a set of pre-defined facial landmark points (e.g., point of corners of eyes, points of the corners of the lip . . . etc.). The output of the Facial Landmark Detection/Tracking 108 contains the locations of the set (e.g., N points) of facial landmark points.
The Face Brightening 110 takes YUV input data and performs adjustment on Y data based on the skin-tone likelihood/score information fed from the analytic module Skin-Tone Detection (STD) 102 to produce brightening effect of the input frame.
The Face Whitening 112 module takes YUV input data and blends the input with a white color map. The white color map is input content-adaptive and is generated within the Face Whitening 112 module. The blending of the input pixels and the white color map is per-pixel wise, adaptive to the Y value of each pixel.
The Skin-Tone-Enhancement (STE) 114 utilizes Skin-Tone Detection (STD) 102 (e.g., VEBox) to perform the saturation enhancement on the skin-tone-color pixels where the enhancement is adaptive to the skin-tone likelihood score.
The Skin Foundation 116 module takes YUV input data and blends the input with a user-selected foundation color where the per-pixel skin-tone likelihood score serves as the blending factor here.
The Skin Smoothing 118 takes YUV input data and adjust all 3-channel information to produce a smooth version of the input.
The Red Lip 120 module takes YUV input data. With the facial landmark information fed into the Red Lip 120 module, the module identifies the lip area of the face if there is a face within the input frame. For input frame with detected faces, Red Lip 120 module further performs color modification for lip area so that a visually pleasant appearance of the users' lips can be obtained.
The Big Eyes 122 module takes YUV input data. With the facial landmark information fed into the Big Eyes 122 module and the users' preference of level of enlargement input from the Application, the Big Eyes 122 module internally derives the proper location within the face and the shape of the eyes users intend to have. Morphological warping is performed following to create the big eyes effect.
The Slim Face 124 module takes YUV input data. With the facial landmark information fed into the Slim Face 124 module and the users' preference of level of slim-face-effect input from the Application, the Slim Face 124 module internally derives the thinner-shape of the original face area and performs morphological warping to create the slim face effect.
The Cute Nose 126 module takes YUV input data. With the facial landmark information fed into the Cute Nose 126 module and the users' preference of level of adjustment input from the Application, the Cute Nose 126 module internally derives the modified shape of the nose area and performs morphological warping to create the narrower/cuter nose effect.
The Happy/Sad 128 module takes YUV input data. With the facial landmark information fed into the Happy/Sad 128 module and the users' preference of level of adjustment input from the Application, the Happy/Sad 128 module internally derives the modified shape of the mouth area and performs morphological warping to create the happy/sad face effect via changing the shape of users' mouths.
The Eye Wrinkles Removal 130 module takes YUV input data. Facial landmark information, rectangle region of the detected face, and Y-channel signal analysis are utilized by the Eye Wrinkles Removal 130 module to locate the area around eyes for processing. Once identifying the target area, a smoothing process is operated on YUV values for pixels within the area to create the wrinkles removal effect.
The Eye Bags Removal 132 module takes YUV input data. Facial landmark information and Y-channel signal analysis are utilized by the Eye Bags Removal 132 module to locate the eye bags regions for processing. Once identifying the regions, a smoothing process is operated on YUV values for pixels within the regions to create the eye bags removal effect.
The Dark Eye Circles Removal 134 block takes YUV input data. Facial landmark information and Y-channel signal analysis are utilized by the Dark Eye Circles Removal 134 to locate the eye bags region for processing. Once identifying the regions, a content-adaptive blending is performed to blend the original YUV values and a pre-defined color value for pixels within the eye bags region. The effect of removing dark eye circles is finally resulted.
The skin-tone enhancement feature 206 shown in
Video codec engine(s) 304 may perform video encoding/decoding. For example, video codec engine(s) 304 may include Decode Engines for video decoding and/or Encode Engines for video encoding.
Video processing engine(s) 306 may perform video pre-post-processing. For example, some parts of video augmentation pipe 100 (e.g. see skin tone detection VEBox-STD 102 and skin tone enhancement VEBox-STE 114 of
Render/3D engine 308 in combination with array of cores 312 may perform rendering, gaming, the like, and/or combinations thereof.
Sampler engine 310 is a separate module (e.g., separate from video processing engine 306) inside GPU 302. For example, sampler engine 310 may be implemented as a hardware module to allow quick sampling access to get the pixels/texels from the original data map, and to allow quick filtering operations.
Often, the type of chips used in servers and mobile devices differ. Accordingly, it may be advantageous to develop server chips that reproduce the functionality of chips for mobile devices. To serve users possessing mobile platforms not equipped with a facial beautification chip, a customized virtual graphics devices may be provide via a cloud-based facial beautification solution. This cloud-based facial beautification solution may have the advantage over power efficiency compared to other cloud-based FB solution which is purely CPU software solution.
In the illustrated example, server Graphic Processing Unit (GPU) 402 may have several available hardware blocks (e.g., independent types of engines) with distinct functionalities. For example, server GPU 402 may include the same and/or similar hardware blocks as mobile platform Graphic Processing Unit (GPU) 302 of
With the emerging of Virtual Desktop Infrastructure (VDI) and general-purpose computing on graphics processing units (GPGPU), GPU virtualization technology may be implemented in cloud computing. In such kind of usage scenarios, one powerful physical GPU 402 on the cloud side may be shared by multiple Virtual Machines (VMs) 412. Note that from a VM's 412 point of view, VMs 412 exclusively owns the virtual graphics device.
Further, as one can see from
The proposed cloud-based facial beautification solution has advantages of both performance and power efficiency compared to others due to two distinct system-wise factors. First, the flexible virtual GPU 400 configuration allows for utilization of GPU instead of CPU; thus better performance may be achieved. Second, the flexible virtual GPU 400 configuration allows for utilization of virtual graphics devices to allow for sharing of the GPU with other workloads; thus a minimum of power leak may be achieved.
As illustrated, step 1402, may operate to derive the affected transform radius R, transform center C, translation vector CM for the nose area in an input image P. For example, affected radius R, transform center C, dislocation vector CM for the nose area may be calculated based at least in part on Facial Landmark Points (of nose area) data as well as a Cute or Bigger Nose Strength setting.
While the Facial Landmark Points (of nose area) is the analytic information consumed by the cute nose filter 1400 at operation 1402, the Cute or Bigger Nose Strength setting is the feature control setting enabled either by an associated Application or by Profile Detection 140 functionality available in the Face Beautification Pipe 100 (see
In some examples, the cute nose filter 1400 at operation 1402 may implement an automatic scheme to determine the morphology of the nose for cute or bigger visual appearance.
At operation 1404, backward warping by translation model may be performed. For example, backward warping by translation model may be performed for each pixel d inside the affected area to get source coordinates s.
At operation 1406, bilinear interpolation may be performed based on the output of operation 1404 and a processed image P′ may be output.
As illustrated, step 1502, may operate to derive the affected transform radius R, transform center C, translation vector CM for the mouth area in an input image P. For example, affected radius R, transform center C, dislocation vector CM for the mouth area may be calculated based at least in part on Facial Landmark Points (of mouth area) data as well as a Happy or Sad Face Strength setting.
While the Facial Landmark Points (of mouth area) is the analytic information consumed by the happy/sad face filter 1500 at step 1502, the Happy or Sad Face Strength setting is the feature control setting enabled by an associated Application. The happy/sad face filter 1500 at step 1502 may implement an automatic scheme to determine the morphology of the mouth for happy or sad appearance.
At operation 1504, backward warping by translation model may be performed. For example, backward warping by translation model may be performed for each pixel d inside the affected area to get source coordinates s.
At operation 1506, bilinear interpolation may be performed based on the output of operation 1504 and a processed image P′ may be output.
As illustrated, step 1602, may operate to generate an eye wrinkles/bags map S. For example, eye wrinkles/bags map S may be generated from luma component of the input image P based at least in part on Facial Landmark Points (of eye and/or nose area) data.
At operation 1604, a smooth version B of input image P may be generated. For example, smooth version B of input image P may be generated via recursive bilateral filtering based at least in part on a Smooth Strength setting.
At operation 1606, a content adaptive blending of input image P and smooth version B may be performed. For example, content adaptive blending of input image P and smooth version B may be performed based at least in part on the generated eye wrinkles/bags map S and a processed image P′ may be output.
In operation, while the Facial Landmark Points (of eye area) is the analytic information consumed by eye bag/wrinkle removal filter 1600 at operation 1602, the Smooth Strength setting is the feature control setting enabled either by an associated Application or by Profile Detection 140 functionality available in the Face Beautification Pipe 100 (see
As illustrated, step 1702, may operate to generate an eye circle map S. For example, eye circle map S may be generated from luma component of the input image P based at least in part on Facial Landmark Points (of eye and/or nose area) data.
At operation 1704, a colored pixels F of input image P may be generated. For example, colored pixels F of input image P may be generated based at least in part on a Desired White Color setting and a Strength setting.
At operation 1706, a content adaptive blending of input image P and colored pixels F may be performed. For example, content adaptive blending of input image P and colored pixels F may be performed based at least in part on the generated eye circle map S and a processed image P′ may be output.
In operation, while the Facial Landmark Points (of eye area) is the analytic information consumed by the dark eye circle removal filter 1700 at step 1702, the Desired White Color & Strength settings are the feature control settings enabled either by an associated Application or by Profile Detection 140 functionality available in the Face Beautification Pipe 100 (see
Process 1800 may be utilized as a computer-implemented method for video processing. Process 1800 may begin at block 1802, “PERFORM FACIAL DETECTION ON THE INPUT VIDEO IMAGES”, where facial detection may be performed. For example, facial detection may be performed via a facial feature logic unit.
Processing may continue from operation 1802 to operation 1804, “PERFORM FACIAL LANDMARK DETECTION AND/OR TRACKING ON THE INPUT VIDEO IMAGES BASED AT LEAST IN PART ON THE FACIAL DETECTION”, where facial landmark detection and/or tracking may be performed. For example, facial landmark detection and/or tracking may be performed based at least in part on the facial detection via the facial feature logic unit.
In some implementations, facial landmark detection may be launched to detect the facial landmark points whenever there is a newly detected face appears in the video. For already detected faces, facial landmark tracking may be applied to update the facial landmark locations from previous frame to the current frame.
Processing may continue from operation 1804 to operation 1806, “MODIFY THE INPUT VIDEO IMAGES BY FACIAL BEAUTIFICATION INTO OUTPUT VIDEO IMAGES”, where facial beautification may be performed. For example, facial beautification may be performed based at least in part on the facial landmark detection via a beautification module.
Processing may continue from operation 1806 to operation 1808, “DISPLAY THE MODIFIED OUTPUT VIDEO IMAGES”, where the modified output video images may be displayed. For example, the modified output video images may be displayed via a display of some sort.
Some additional and/or alternative details related to process 1800 may be illustrated in one or more examples of implementations discussed in greater detail below with regard to
By way of non-limiting example, process 1900 will be described herein with reference to example video augmentation system 2000 including video augmentation system pipe 100 of
Process 1900 may begin at operation 1912, “RECEIVE INPUT VIDEO IMAGES”, where input video images may be received. For example, input video images may be received via the skin tone logic unit 202, the facial feature logic unit 204, and/or the beautification module 206/208.
Process 1900 may continue at operation 1914, “PERFORM SKIN TONE DETECTION”, where a skin tone likelihood score may be determined. For example, skin tone likelihood score may be determined from the input video images via skin tone logic unit 202.
Process 1900 may continue at operation 1916, “SMOOTH SKIN TONE SCORE”, where the skin tone likelihood score may be smoothed. For example, the skin tone likelihood score may be smoothed via the skin tone logic unit 202.
In some implementations, skin tone likelihood score 1918 from skin tone logic unit 202 may be output to skin tone beatification module 206.
Process 1900 may continue at operation 1920, “FACE DETECTION”, where facial detection may be performed. For example, facial detection may be performed on the input video images via the facial feature logic unit 204.
Process 1900 may continue at operation 1922, “LANDMARK DETECTION AND/OR TRACKING”, where facial landmark detection and/or tracking may be performed. For example, facial landmark detection and/or tracking may be performed on the input video images based at least in part on the facial detection via the facial feature logic unit 204.
In some implementations, landmark points data 1924 from facial feature logic unit 204 may be output to landmark beatification module 208.
In some implementations, detected face data 1926 from facial feature logic unit 204 may be output to profile detection module 140.
Process 1900 may continue at operation 1928, “FACIAL RECOGNITION PROFILE DETECTION”, where facial recognition profile detection may be performed. For example, facial recognition profile detection may be performed via profile detection module 140 based at least in part on the face detection.
In some implementations, strength settings 1930 from profile detection module 140 may be output to skin tone beatification module 206 and or to landmark beatification module 208.
For example, profile detection module 140 may modify strength settings associated with various facial beatification operations based at least in part on the face detection. For example, profile detection module 140 may modify the strength settings based at least in part on one or more of the following detected facial attributes: gender, age, and race.
Alternatively, in some implementations, strength settings 1930 may instead be preset and/or be specified by a user's preference and may be output to skin tone beatification module 206 and or to landmark beatification module 208.
Process 1900 may continue at operation 1932, “BRIGHTENING”; operation 1934, “WHITENING”; operation 1936, “ENHANCEMENT”; operation 1938, “FOUNDATION”; and operation 1940, “SMOOTHING”.
As will be described below (at one or more of operations 1932, 1934, 1936, 1938, 1940, 1942, 1944, 1946, 1948, 1950, 1952, 1954, and 1956) beautification module 206/208 may modify the input video images by facial beautification into output video images based at least in part on the skin tone likelihood score and/or the facial landmark detection.
For example, one or more of the following skin tone beautification operations (face brightening, face whitening, skin tone enhancement, skin foundation, and skin smoothing) may be performed via skin tone beautification portion 206 of the beautification module 204/206 based at least in part on the strength settings.
Similarly, and performing one or more of the following skin tone beautification operations (skin tone enhancement, skin foundation, and skin smoothing) may be performed via skin tone beautification portion 206 of the beautification module 204/206 based at least in part on the skin tone likelihood score.
In some implementations, operations 1932, 1934, 1936, 1938, and 1940 may be done sequentially, so that the output of one operation (e.g., operation 1932) may be used as the input video image to start the next operation (e.g., operation 1934). The above examples may include specific combination of features. However, such the above examples are not limited in this regard and, in various implementations, the above examples may include the undertaking only a subset of such features, undertaking a different order of such features, undertaking a different combination of such features, and/or undertaking additional features than those features explicitly listed.
In some implementations, output video images 1941 modified by skin tone beautification may be output to landmark beatification module 208.
Process 1900 may continue at operation 1942, “RED LIP”; operation 1944, “BIG EYES”; operation 1946, “SLIM FACE”; operation 1948, “CUTE NOSE”; operation 1950, “HAPPY/SAD FACE”; operation 1952, “EYE WRINKLE REMOVA”; operation 1954, “EYE BAGS REMOVAL”; and operation 1956, “DARK EYE CIRCLES REMOVAL”.
For example, one or more of the following landmark beautification operations (red lip alteration, big eye alteration, slim face alteration, cute/big nose alteration, happy/sad face alteration, eye wrinkle removal, eye bags removal, and dark eye circles removal) may be performed via a landmark beautification portion 208 of the beautification module 206/208 based at least in part on the strength settings and the landmark points.
In some implementations, operations 1942, 1944, 1946, 1948, 1950, 1952, 1954, and 1956 may be done sequentially, so that the output of one operation (e.g., operation 1952) may be used as the input video image to start the next operation (e.g., operation 1954). The above examples may include specific combination of features. However, such the above examples are not limited in this regard and, in various implementations, the above examples may include the undertaking only a subset of such features, undertaking a different order of such features, undertaking a different combination of such features, and/or undertaking additional features than those features explicitly listed.
In some implementations, operation 1946 may include performing, via a cute nose filter portion of the beautification module, the following operations: 1) deriving an affected area transform radius R, transform center C, and translation vector CM for an associated nose area in an input image P based at least in part on the landmark points as well as on a cute or bigger nose strength setting; 2) performing backward warping by translation model for each pixel d inside the affected area to get source coordinates s; and 3) performing bilinear interpolation based at least in part on the backward warping of the affected area to output a processed image P′.
In some implementations, operation 1948 may include performing, via a happy/sad face filter portion of the beautification module, the following operations: 1) deriving an affected area transform radius R, transform center C, and translation vector CM for the mouth area in an input image P based at least in part on the landmark points associated with a mouth area as well as a happy/sad face strength setting; 2) performing backward warping by translation model for each pixel d inside the affected area to get source coordinates s; and 3) performing bilinear interpolation based at least in part on the backward warping of the affected area to output a processed image P′.
In some implementations, operation 1952 and/or 1954 may include performing, via a eye bag/wrinkle removal filter portion of the beautification module, the following operations: 1) generating an eye wrinkles/bags map S from luma component of an input image P based at least in part on landmark points associated with an eye and/or nose area; 2) generating a smooth version B of input image P via a recursive bilateral filtering based at least in part on a smooth strength setting; and 3) performing a content adaptive blending of input image P and smooth version B based at least in part on the generated eye wrinkles/bags map S to output a processed image P′.
In some implementations, operation 1956 may include performing, via a dark eye circle removal filter portion of the beautification module, the following operations: 1) generating an eye circle map S from luma component of an input image P based at least in part on landmark points associated with an eye and/or nose area; 2) generating colored pixels F of input image P may be generated. For example, colored pixels F of input image P may be generated based at least in part on a desired white color setting and a strength setting; and 3) performing a content adaptive blending of input image P and colored pixels F based at least in part on the generated eye circle map S to output a processed image P′.
In some implementations, output video images 1958 modified by landmark beautification may be output to display 141.
Alternatively, in server-type implementations, such output video images 1958 modified by landmark beautification might be output for transmission to one or more remote devices (not illustrated here).
Process 1900 may continue at operation 1960, “DISPLAY MODIFIED OUTPUT VIDEO IMAGES”, where the modified output video images may be displayed. For example, the modified output video images as modified by the beautification module 206/208 may be displayed via display 141.
In some implementations, video augmentation pipe 100 may be implemented on a mobile platform type GPU.
Alternatively, in some implementations, video augmentation pipe 100 may be implemented on a server platform type GPU. In such an implementation, the server platform type GPU may include several virtual GPUs that may share one physical GPU. The server platform type GPU may be configured to transfer input video images as well as output video images modified by the facial beatification operations between one or more remote devices (not shown) in communication with sever platform GPU.
Some additional and/or alternative details related to process 1900 and other processes discussed herein may be illustrated in one or more examples of implementations discussed herein and, in particular, with respect to
Various components of the systems and/or processes described herein may be implemented in software, firmware, and/or hardware and/or any combination thereof. For example, various components of the systems and/or processes described herein may be provided, at least in part, by hardware of a computing System-on-a-Chip (SoC) such as may be found in a computing system such as, for example, a smart phone. Those skilled in the art may recognize that systems described herein may include additional components that have not been depicted in the corresponding figures.
As used in any implementation described herein, the term “module” may refer to a “component” or to a “logic unit”, as these terms are described below. Accordingly, the term “module” may refer to any combination of software logic, firmware logic, and/or hardware logic configured to provide the functionality described herein. For example, one of ordinary skill in the art will appreciate that operations performed by hardware and/or firmware may alternatively be implemented via a software component, which may be embodied as a software package, code and/or instruction set, and also appreciate that a logic unit may also utilize a portion of software to implement its functionality.
As used in any implementation described herein, the term “component” refers to any combination of software logic and/or firmware logic configured to provide the functionality described herein. The software logic may be embodied as a software package, code and/or instruction set, and/or firmware that stores instructions executed by programmable circuitry. The components may, collectively or individually, be embodied for implementation as part of a larger system, for example, an integrated circuit (IC), system on-chip (SoC), and so forth.
As used in any implementation described herein, the term “logic unit” refers to any combination of firmware logic and/or hardware logic configured to provide the functionality described herein. The “hardware”, as used in any implementation described herein, may include, for example, singly or in any combination, hardwired circuitry, programmable circuitry, state machine circuitry, and/or firmware that stores instructions executed by programmable circuitry. The logic units may, collectively or individually, be embodied as circuitry that forms part of a larger system, for example, an integrated circuit (IC), system on-chip (SoC), and so forth. For example, a logic unit may be embodied in logic circuitry for the implementation firmware or hardware of the systems discussed herein. Further, one of ordinary skill in the art will appreciate that operations performed by hardware and/or firmware may also utilize a portion of software to implement the functionality of the logic unit.
In addition, any one or more of the blocks of the processes described herein may be undertaken in response to instructions provided by one or more computer program products. Such program products may include signal bearing media providing instructions that, when executed by, for example, a processor, may provide the functionality described herein. The computer program products may be provided in any form of computer readable medium. Thus, for example, a processor including one or more processor core(s) may undertake one or more operations in response to instructions conveyed to the processor by a computer readable medium.
As illustrated, processors 2002, memory store 2004, GPU 2006, display 2008, coder 2012, and/or antenna 2014 may be capable of communication with one another and/or communication with portions of logic modules 2010.
In some implementations, graphics processing system 2000 may include antenna 2012. For example, antenna 2012 may be configured to transmit or receive an encoded bitstream of video data, for example. Processor(s) 2002 and/or GPU(s) 1006 may be any type of processor and/or processing unit. For example, processor(s) 2002 may include distinct central processing units, distinct graphic processing units, integrated system-on-a-chip (SoC) architectures, the like, and/or combinations thereof. In addition, memory store(s) 2008 may be any type of memory. For example, memory store(s) 2004 may be volatile memory (e.g., Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), etc.) or non-volatile memory (e.g., flash memory, etc.), and so forth. In a non-limiting example, memory store(s) 2004 may be implemented by cache memory. Further, in some implementations, graphics processing system 2000 may include display device 208. Display device 2008 may be configured to present video data as images 2009.
In some implementations, logic modules 2010 may embody various modules as discussed with respect to any system or subsystem described herein. In various embodiments, some of logic modules 2010 may be implemented in hardware, while software may implement other logic modules. For example, in some embodiments, some of logic modules 2010 may be implemented by application-specific integrated circuit (ASIC) logic while other logic modules may be provided by software instructions executed by logic such as GPU 2006, for example. However, the present disclosure is not limited in this regard and some of logic modules 2010 may be implemented by any combination of hardware, firmware and/or software.
For example, logic modules 2010 may include video augmentation pipe 100 and/or the like configured to implement operations of one or more of the implementations described herein.
In various implementations, system 2100 includes a platform 2102 coupled to a display 2120. Platform 2102 may receive content from a content device such as content services device(s) 2130 or content delivery device(s) 2140 or other similar content sources. A navigation controller 2150 including one or more navigation features may be used to interact with, for example, platform 2102 and/or display 2120. Each of these components is described in greater detail below.
In various implementations, platform 2102 may include any combination of a chipset 2105, processor 2110, memory 2112, antenna 2113, storage 2114, graphics subsystem 2115, applications 2116 and/or radio 2118. Chipset 2105 may provide intercommunication among processor 2110, memory 2112, storage 2114, graphics subsystem 2115, applications 2116 and/or radio 2118. For example, chipset 2105 may include a storage adapter (not depicted) capable of providing intercommunication with storage 2114.
Processor 2110 may be implemented as a Complex Instruction Set Computer (CISC) or Reduced Instruction Set Computer (RISC) processors, x86 instruction set compatible processors, multi-core, or any other microprocessor or central processing unit (CPU). In various implementations, processor 2110 may be dual-core processor(s), dual-core mobile processor(s), and so forth.
Memory 2112 may be implemented as a volatile memory device such as, but not limited to, a Random Access Memory (RAM), Dynamic Random Access Memory (DRAM), or Static RAM (SRAM).
Storage 2114 may be implemented as a non-volatile storage device such as, but not limited to, a magnetic disk drive, optical disk drive, tape drive, an internal storage device, an attached storage device, flash memory, battery backed-up SDRAM (synchronous DRAM), and/or a network accessible storage device. In various implementations, storage 2114 may include technology to increase the storage performance enhanced protection for valuable digital media when multiple hard drives are included, for example.
Graphics subsystem 2115 may perform processing of images such as still or video for display. Graphics subsystem 2115 may be a graphics processing unit (GPU) or a visual processing unit (VPU), for example. An analog or digital interface may be used to communicatively couple graphics subsystem 2115 and display 2120. For example, the interface may be any of a High-Definition Multimedia Interface, DisplayPort, wireless HDMI, and/or wireless HD compliant techniques. Graphics subsystem 2115 may be integrated into processor 2110 or chipset 2105. In some implementations, graphics subsystem 2115 may be a stand-alone device communicatively coupled to chipset 2105.
The graphics and/or video processing techniques described herein may be implemented in various hardware architectures. For example, graphics and/or video functionality may be integrated within a chipset. Alternatively, a discrete graphics and/or video processor may be used. As still another implementation, the graphics and/or video functions may be provided by a general purpose processor, including a multi-core processor. In further embodiments, the functions may be implemented in a consumer electronics device.
Radio 2118 may include one or more radios capable of transmitting and receiving signals using various suitable wireless communications techniques. Such techniques may involve communications across one or more wireless networks. Example wireless networks include (but are not limited to) wireless local area networks (WLANs), wireless personal area networks (WPANs), wireless metropolitan area network (WMANs), cellular networks, and satellite networks. In communicating across such networks, radio 2118 may operate in accordance with one or more applicable standards in any version.
In various implementations, display 2120 may include any television type monitor or display. Display 2120 may include, for example, a computer display screen, touch screen display, video monitor, television-like device, and/or a television. Display 2120 may be digital and/or analog. In various implementations, display 2120 may be a holographic display. Also, display 2120 may be a transparent surface that may receive a visual projection. Such projections may convey various forms of information, images, and/or objects. For example, such projections may be a visual overlay for a mobile augmented reality (MAR) application. Under the control of one or more software applications 2116, platform 2102 may display user interface 2122 on display 2120.
In various implementations, content services device(s) 2130 may be hosted by any national, international and/or independent service and thus accessible to platform 2102 via the Internet, for example. Content services device(s) 2130 may be coupled to platform 2102 and/or to display 2120. Platform 2102 and/or content services device(s) 2130 may be coupled to a network 2160 to communicate (e.g., send and/or receive) media information to and from network 2160. Content delivery device(s) 2140 also may be coupled to platform 2102 and/or to display 2120.
In various implementations, content services device(s) 2130 may include a cable television box, personal computer, network, telephone, Internet enabled devices or appliance capable of delivering digital information and/or content, and any other similar device capable of unidirectionally or bidirectionally communicating content between content providers and platform 2102 and/display 2120, via network 2160 or directly. It will be appreciated that the content may be communicated unidirectionally and/or bidirectionally to and from any one of the components in system 2100 and a content provider via network 2160. Examples of content may include any media information including, for example, video, music, medical and gaming information, and so forth.
Content services device(s) 2130 may receive content such as cable television programming including media information, digital information, and/or other content. Examples of content providers may include any cable or satellite television or radio or Internet content providers. The provided examples are not meant to limit implementations in accordance with the present disclosure in any way.
In various implementations, platform 2102 may receive control signals from navigation controller 2150 having one or more navigation features. The navigation features of controller 2150 may be used to interact with user interface 2122, for example. In various embodiments, navigation controller 2150 may be a pointing device that may be a computer hardware component (specifically, a human interface device) that allows a user to input spatial (e.g., continuous and multi-dimensional) data into a computer. Many systems such as graphical user interfaces (GUI), and televisions and monitors allow the user to control and provide data to the computer or television using physical gestures.
Movements of the navigation features of controller 2150 may be replicated on a display (e.g., display 2120) by movements of a pointer, cursor, focus ring, or other visual indicators displayed on the display. For example, under the control of software applications 2116, the navigation features located on navigation controller 2150 may be mapped to virtual navigation features displayed on user interface 2122. In various embodiments, controller 2150 may not be a separate component but may be integrated into platform 2102 and/or display 2120. The present disclosure, however, is not limited to the elements or in the context shown or described herein.
In various implementations, drivers (not shown) may include technology to enable users to instantly turn on and off platform 2102 like a television with the touch of a button after initial boot-up, when enabled, for example. Program logic may allow platform 2102 to stream content to media adaptors or other content services device(s) 2130 or content delivery device(s) 2140 even when the platform is turned “off.” In addition, chipset 2105 may include hardware and/or software support for (5.1) surround sound audio and/or high definition (7.1) surround sound audio, for example. Drivers may include a graphics driver for integrated graphics platforms. In various embodiments, the graphics driver may comprise a peripheral component interconnect (PCI) Express graphics card.
In various implementations, any one or more of the components shown in system 2100 may be integrated. For example, platform 2102 and content services device(s) 2130 may be integrated, or platform 2102 and content delivery device(s) 2140 may be integrated, or platform 2102, content services device(s) 2130, and content delivery device(s) 2140 may be integrated, for example. In various embodiments, platform 2102 and display 2120 may be an integrated unit. Display 2120 and content service device(s) 2130 may be integrated, or display 2120 and content delivery device(s) 2140 may be integrated, for example. These examples are not meant to limit the present disclosure.
In various embodiments, system 2100 may be implemented as a wireless system, a wired system, or a combination of both. When implemented as a wireless system, system 2100 may include components and interfaces suitable for communicating over a wireless shared media, such as one or more antennas, transmitters, receivers, transceivers, amplifiers, filters, control logic, and so forth. An example of wireless shared media may include portions of a wireless spectrum, such as the RF spectrum and so forth. When implemented as a wired system, system 2100 may include components and interfaces suitable for communicating over wired communications media, such as input/output (I/O) adapters, physical connectors to connect the I/O adapter with a corresponding wired communications medium, a network interface card (NIC), disc controller, video controller, audio controller, and the like. Examples of wired communications media may include a wire, cable, metal leads, printed circuit board (PCB), backplane, switch fabric, semiconductor material, twisted-pair wire, co-axial cable, fiber optics, and so forth.
Platform 2102 may establish one or more logical or physical channels to communicate information. The information may include media information and control information. Media information may refer to any data representing content meant for a user. Examples of content may include, for example, data from a voice conversation, videoconference, streaming video, electronic mail (“email”) message, voice mail message, alphanumeric symbols, graphics, image, video, text and so forth. Data from a voice conversation may be, for example, speech information, silence periods, background noise, comfort noise, tones and so forth. Control information may refer to any data representing commands, instructions or control words meant for an automated system. For example, control information may be used to route media information through a system, or instruct a node to process the media information in a predetermined manner. The embodiments, however, are not limited to the elements or in the context shown or described in
As described above, system 2100 may be embodied in varying physical styles or form factors.
As described above, examples of a mobile computing device may include a personal computer (PC), laptop computer, ultra-laptop computer, tablet, touch pad, portable computer, handheld computer, palmtop computer, personal digital assistant (PDA), cellular telephone, combination cellular telephone/PDA, television, smart device (e.g., smart phone, smart tablet or smart television), mobile internet device (MID), messaging device, data communication device, cameras (e.g. point-and-shoot cameras, super-zoom cameras, digital single-lens reflex (DSLR) cameras), and so forth.
Examples of a mobile computing device also may include computers that are arranged to be worn by a person, such as a wrist computer, finger computer, ring computer, eyeglass computer, belt-clip computer, arm-band computer, shoe computers, clothing computers, and other wearable computers. In various embodiments, for example, a mobile computing device may be implemented as a smart phone capable of executing computer applications, as well as voice communications and/or data communications. Although some embodiments may be described with a mobile computing device implemented as a smart phone by way of example, it may be appreciated that other embodiments may be implemented using other wireless mobile computing devices as well. The embodiments are not limited in this context.
As shown in
Various embodiments may be implemented using hardware elements, software elements, or a combination of both. Examples of hardware elements may include processors, microprocessors, circuits, circuit elements (e.g., transistors, resistors, capacitors, inductors, and so forth), integrated circuits, application specific integrated circuits (ASIC), programmable logic devices (PLD), digital signal processors (DSP), field programmable gate array (FPGA), logic gates, registers, semiconductor device, chips, microchips, chip sets, and so forth. Examples of software may include software components, programs, applications, computer programs, application programs, system programs, machine programs, operating system software, middleware, firmware, software modules, routines, subroutines, functions, methods, procedures, software interfaces, application program interfaces (API), instruction sets, computing code, computer code, code segments, computer code segments, words, values, symbols, or any combination thereof. Determining whether an embodiment is implemented using hardware elements and/or software elements may vary in accordance with any number of factors, such as desired computational rate, power levels, heat tolerances, processing cycle budget, input data rates, output data rates, memory resources, data bus speeds and other design or performance constraints.
One or more aspects of at least one embodiment may be implemented by representative instructions stored on a machine-readable medium which represents various logic within the processor, which when read by a machine causes the machine to fabricate logic to perform the techniques described herein. Such representations, known as “IP cores” may be stored on a tangible, machine readable medium and supplied to various customers or manufacturing facilities to load into the fabrication machines that actually make the logic or processor.
While certain features set forth herein have been described with reference to various implementations, this description is not intended to be construed in a limiting sense. Hence, various modifications of the implementations described herein, as well as other implementations, which are apparent to persons skilled in the art to which the present disclosure pertains are deemed to lie within the spirit and scope of the present disclosure.
The following examples pertain to further embodiments.
In one example, a computer-implemented method for video augmentation may include performing, via a facial feature logic unit of the GPU, facial detection on the input video images; performing, via the facial feature logic unit, facial landmark point detection and/or tracking on the input video images based at least in part on the facial detection; and modifying, via a beautification module of the GPU, the input video images by facial beautification into output video images based at least in part on the facial landmark point detection.
In another example, the computer-implemented method for video augmentation may include receiving input video images via a skin tone logic unit, the facial feature logic unit, and the beautification module; determining, via the skin tone logic unit, a skin tone likelihood score on input video images; smoothing, via the skin tone logic unit, the skin tone likelihood score; performing, via a profile detection module, facial recognition profile detection based at least in part on the face detection; modifying, via the profile detection module, strength settings associated with various facial beatification operations based at least in part on the face detection, where the strength settings may be based at least in part on one or more of the following detected facial attributes: gender, age, and race; performing, via a skin tone beautification portion of the beautification module, one or more of the following skin tone beautification operations based at least in part on the strength settings: face brightening, face whitening, skin tone enhancement, skin foundation, and skin smoothing, and performing one or more of the following skin tone beautification operations based at least in part on the skin tone likelihood score: skin tone enhancement, skin foundation, and skin smoothing; performing, via a landmark beautification portion of the beautification module, one or more of the following landmark beautification operations based at least in part on the strength settings and the landmark points as well as the performed skin tone beautification operations: red lip alteration, big eye alteration, and slim face alteration; performing, via a cute nose filter portion of the beautification module, the following operations: deriving an affected area transform radius R, transform center C, and translation vector CM for an associated nose area in an input image P based at least in part on the landmark points as well as on a cute or bigger nose strength setting; performing backward warping by translation model for each pixel d inside the affected area to get source coordinates s; performing bilinear interpolation based at least in part on the backward warping of the affected area to output a processed image P′; performing, via a happy/sad face filter portion of the beautification module, the following operations: deriving an affected area transform radius R, transform center C, and translation vector CM for the mouth area in an input image P based at least in part on the landmark points associated with a mouth area as well as a happy/sad face strength setting; performing backward warping by translation model for each pixel d inside the affected area to get source coordinates s; performing bilinear interpolation based at least in part on the backward warping of the affected area to output a processed image P′; performing, via a eye bag/wrinkle removal filter portion of the beautification module, the following operations: generating an eye wrinkles/bags map S from luma component of an input image P based at least in part on landmark points associated with an eye and/or nose area; generating a smooth version B of input image P via a recursive bilateral filtering based at least in part on a smooth strength setting; performing a content adaptive blending of input image P and smooth version B based at least in part on the generated eye wrinkles/bags map S to output a processed image P′; performing, via a dark eye circle removal filter portion of the beautification module, the following operations: generating an eye circle map S from luma component of an input image P based at least in part on landmark points associated with an eye and/or nose area; generating colored pixels F of input image P may be generated. For example, colored pixels F of input image P may be generated based at least in part on a desired white color setting and a strength setting; and performing a content adaptive blending of input image P and colored pixels F based at least in part on the generated eye circle map S to output a processed image P′; and displaying, via a display, output video images as modified by the beautification module.
In other examples, a mobile computing device for video augmentation may include one or more graphics processing units, the one or more graphics processing units including a facial feature logic unit and a beautification module; one or more memory stores communicatively coupled to the one or more graphics processing units; where the one or more graphics processing units are configured to: perform, via the facial feature logic unit, facial detection on the input video images; perform, via the facial feature logic unit, facial landmark point detection and/or tracking on the input video images based at least in part on the facial detection; and modify, via the beautification module, the input video images by facial beautification into output video images based at least in part on the facial landmark point detection.
In another example, the mobile computing device for video augmentation may further include the one or more graphics processing units being further configured to: receive input video images via a skin tone logic unit, the facial feature logic unit, and the beautification module; determine, via the skin tone logic unit, a skin tone likelihood score on input video images; smooth, via the skin tone logic unit, the skin tone likelihood score; perform, via a profile detection module, facial recognition profile detection based at least in part on the face detection; modify, via the profile detection module, strength settings associated with various facial beatification operations based at least in part on the face detection, where the strength settings may be based at least in part on one or more of the following detected facial attributes: gender, age, and race; perform, via a skin tone beautification portion of the beautification module, one or more of the following skin tone beautification operations based at least in part on the strength settings: face brightening, face whitening, skin tone enhancement, skin foundation, and skin smoothing, and performing one or more of the following skin tone beautification operations based at least in part on the skin tone likelihood score: skin tone enhancement, skin foundation, and skin smoothing; perform, via a landmark beautification portion of the beautification module, one or more of the following landmark beautification operations based at least in part on the strength settings and the landmark points as well as the performed skin tone beautification operations: red lip alteration, big eye alteration, and slim face alteration; perform, via a cute nose filter portion of the beautification module, the following operations: derive an affected area transform radius R, transform center C, and translation vector CM for an associated nose area in an input image P based at least in part on the landmark points as well as on a cute or bigger nose strength setting; perform backward warping by translation model for each pixel d inside the affected area to get source coordinates s; perform bilinear interpolation based at least in part on the backward warping of the affected area to output a processed image P′; perform, via a happy/sad face filter portion of the beautification module, the following operations: derive an affected area transform radius R, transform center C, and translation vector CM for the mouth area in an input image P based at least in part on the landmark points associated with a mouth area as well as a happy/sad face strength setting; perform backward warping by translation model for each pixel d inside the affected area to get source coordinates s; perform bilinear interpolation based at least in part on the backward warping of the affected area to output a processed image P′; perform, via a eye bag/wrinkle removal filter portion of the beautification module, the following operations: generate an eye wrinkles/bags map S from luma component of an input image P based at least in part on landmark points associated with an eye and/or nose area; generate a smooth version B of input image P via a recursive bilateral filtering based at least in part on a smooth strength setting; perform a content adaptive blending of input image P and smooth version B based at least in part on the generated eye wrinkles/bags map S to output a processed image P′; perform, via a dark eye circle removal filter portion of the beautification module, the following operations: generate an eye circle map S from a luma component of an input image P based at least in part on landmark points associated with an eye and/or nose area; generate colored pixels F of input image P may be generated. For example, colored pixels F of input image P may be generated based at least in part on a desired white color setting and a strength setting; and perform a content adaptive blending of input image P and colored pixels F based at least in part on the generated eye circle map S to output a processed image P′; and the mobile computing device further including a display, the display configured to display output video images as modified by the beautification module.
In other examples, an apparatus for video augmentation on a computer may include a graphics processing unit (GPU), the graphics processing unit configured to: perform, via a facial feature logic unit, facial detection on the input video images; perform, via the facial feature logic unit, facial landmark point detection and/or tracking on the input video images based at least in part on the facial detection; and modify, via a beautification module, the input video images by facial beautification into output video images based at least in part on the facial landmark point detection.
In another example, the apparatus for video augmentation may further include the graphics processing unit being further configured to: receive input video images via a skin tone logic unit, the facial feature logic unit, and the beautification module; determine, via the skin tone logic unit, a skin tone likelihood score on input video images; smooth, via the skin tone logic unit, the skin tone likelihood score; perform, via a profile detection module, facial recognition profile detection based at least in part on the face detection; modify, via the profile detection module, strength settings associated with various facial beatification operations based at least in part on the face detection, where the strength settings may be based at least in part on one or more of the following detected facial attributes: gender, age, and race; perform, via a skin tone beautification portion of the beautification module, one or more of the following skin tone beautification operations based at least in part on the strength settings: face brightening, face whitening, skin tone enhancement, skin foundation, and skin smoothing, and performing one or more of the following skin tone beautification operations based at least in part on the skin tone likelihood score: skin tone enhancement, skin foundation, and skin smoothing; perform, via a landmark beautification portion of the beautification module, one or more of the following landmark beautification operations based at least in part on the strength settings and the landmark points as well as the performed skin tone beautification operations: red lip alteration, big eye alteration, and slim face alteration; perform, via a cute nose filter portion of the beautification module, the following operations: derive an affected area transform radius R, transform center C, and translation vector CM for an associated nose area in an input image P based at least in part on the landmark points as well as on a cute or bigger nose strength setting; perform backward warping by translation model for each pixel d inside the affected area to get source coordinates s; perform bilinear interpolation based at least in part on the backward warping of the affected area to output a processed image P′; perform, via a happy/sad face filter portion of the beautification module, the following operations: derive an affected area transform radius R, transform center C, and translation vector CM for the mouth area in an input image P based at least in part on the landmark points associated with a mouth area as well as a happy/sad face strength setting; perform backward warping by translation model for each pixel d inside the affected area to get source coordinates s; perform bilinear interpolation based at least in part on the backward warping of the affected area to output a processed image P′; perform, via a eye bag/wrinkle removal filter portion of the beautification module, the following operations: generate an eye wrinkles/bags map S from luma component of an input image P based at least in part on landmark points associated with an eye and/or nose area; generate a smooth version B of input image P via a recursive bilateral filtering based at least in part on a smooth strength setting; perform a content adaptive blending of input image P and smooth version B based at least in part on the generated eye wrinkles/bags map S to output a processed image P′; perform, via a dark eye circle removal filter portion of the beautification module, the following operations: generate an eye circle map S from a luma component of an input image P based at least in part on landmark points associated with an eye and/or nose area; generate colored pixels F of input image P may be generated. For example, colored pixels F of input image P may be generated based at least in part on a desired white color setting and a strength setting; and perform a content adaptive blending of input image P and colored pixels F based at least in part on the generated eye circle map S to output a processed image P′; and cause to be displayed, via a display, output video images as modified by the beautification module.
In a further example, at least one machine readable medium may include a plurality of instructions that in response to being executed on a computing device, causes the computing device to perform the method according to any one of the above examples.
In a still further example, an apparatus may include means for performing the methods according to any one of the above examples.
The above examples may include specific combination of features. However, such the above examples are not limited in this regard and, in various implementations, the above examples may include the undertaking only a subset of such features, undertaking a different order of such features, undertaking a different combination of such features, and/or undertaking additional features than those features explicitly listed. For example, all features described with respect to the example methods may be implemented with respect to the example apparatus, the example systems, and/or the example articles, and vice versa.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/CN2015/089441 | 9/11/2015 | WO | 00 |
| Publishing Document | Publishing Date | Country | Kind |
|---|---|---|---|
| WO2017/041295 | 3/16/2017 | WO | A |
| Number | Name | Date | Kind |
|---|---|---|---|
| 9449412 | Rogers | Sep 2016 | B1 |
| 9495582 | Guissin | Nov 2016 | B2 |
| 20100278426 | Piramuthu | Nov 2010 | A1 |
| 20110234611 | Singhal | Sep 2011 | A1 |
| 20120008002 | Bigioi | Jan 2012 | A1 |
| 20130216154 | Li | Aug 2013 | A1 |
| 20140147003 | Li et al. | May 2014 | A1 |
| 20140176548 | Green | Jun 2014 | A1 |
| 20140369554 | Albuz | Dec 2014 | A1 |
| 20150169938 | Yao et al. | Jun 2015 | A1 |
| 20160042224 | Liu | Feb 2016 | A1 |
| 20160086355 | Zhang | Mar 2016 | A1 |
| Number | Date | Country |
|---|---|---|
| 103180873 | Jun 2013 | CN |
| 104537612 | Apr 2015 | CN |
| Entry |
|---|
| Yoda, Akira, Yukihiro Iguchi, and Kaoru Arakawa. “Development of nonlinear filter bank system for real-time beautification of facial video using GPGPU.” Communications and Information Technologies (ISCIT), 2010 International Symposium on. IEEE, 2010. |
| Xu, Lin, Yangzhou Du, and Yimin Zhang. “An automatic framework for example-based virtual makeup.” Image Processing (ICIP), 2013 20th IEEE International Conference on. IEEE, 2013. |
| Liang, Lingyu, Lianwen Jin, and Xuelong Li. “Facial skin beautification using adaptive region-aware masks.” IEEE transactions on cybernetics 44.12 (2014): 2600-2612. |
| Leyvand, Tommer, et al. “Data-driven enhancement of facial attractiveness.” ACM Transactions on Graphics (TOG). vol. 27. No. 3. ACM, 2008. |
| Gao, Shelley, Christopher Werner, and Amy A. Gooch. “Morphable guidelines for the human head.” Proceedings of the Symposium on Computational Aesthetics. ACM, 2013. |
| Zhao, Yajie, et al. “Video face beautification.” Multimedia and Expo (ICME), 2014 IEEE International Conference on. IEEE, 2014. (Year: 2014). |
| International Search Report for International Patent Application No. PCT/CN2015/089441, dated Jun. 1, 2016, 3 pages. |
| Written Opinion for International Patent Application No. PCT/CN2015/089441, dated Jun. 1, 2016, 5 pages. |
| International Preliminary Report on Patentability for International Patent Application No. PCT/CN2015/089441, dated Mar. 22, 2018, 7 pages. |
| Number | Date | Country | |
|---|---|---|---|
| 20170262970 A1 | Sep 2017 | US |
