Patent application number | Description | Published |
20100194872 | BODY SCAN - A depth image of a scene may be received, observed, or captured by a device. The depth image may then be analyzed to determine whether the depth image includes a human target. For example, the depth image may include one or more targets including a human target and non-human targets. Each of the targets may be flood filled and compared to a pattern to determine whether the target may be a human target. If one or more of the targets in the depth image includes a human target, the human target may be scanned. A skeletal model of the human target may then be generated based on the scan. | 08-05-2010 |
20100195867 | VISUAL TARGET TRACKING USING MODEL FITTING AND EXEMPLAR - A method of tracking a target includes receiving an observed depth image of the target from a source and analyzing the observed depth image with a prior-trained collection of known poses to find an exemplar pose that represents an observed pose of the target. The method further includes rasterizing a model of the target into a synthesized depth image having a rasterized pose and adjusting the rasterized pose of the model into a model-fitting pose based, at least in part, on differences between the observed depth image and the synthesized depth image. Either the exemplar pose or the model-fitting pose is then selected to represent the target. | 08-05-2010 |
20100278431 | Systems And Methods For Detecting A Tilt Angle From A Depth Image - A depth image of a scene may be received, observed, or captured by a device. A human target in the depth image may then be scanned for one or more body parts such as shoulders, hips, knees, or the like. A tilt angle may then be calculated based on the body parts. For example, a first portion of pixels associated with an upper body part such as the shoulders and a second portion of pixels associated with a lower body part such as a midpoint between the hips and knees may be selected. The tilt angle may then be calculated using the first and second portions of pixels. | 11-04-2010 |
20100302395 | Environment And/Or Target Segmentation - A depth image of a scene may be observed or captured by a capture device. The depth image may include a human target and an environment. One or more pixels of the depth image may be analyzed to determine whether the pixels in the depth image are associated with the environment of the depth image. The one or more pixels associated with the environment may then be discarded to isolate the human target and the depth image with the isolated human target may be processed. | 12-02-2010 |
20110032336 | BODY SCAN - A depth image of a scene may be received, observed, or captured by a device. The depth image may then be analyzed to determine whether the depth image includes a human target. For example, the depth image may include one or more targets including a human target and non-human targets. Each of the targets may be flood filled and compared to a pattern to determine whether the target may be a human target. If one or more of the targets in the depth image includes a human target, the human target may be scanned. A skeletal model of the human target may then be generated based on the scan. | 02-10-2011 |
20110058709 | VISUAL TARGET TRACKING USING MODEL FITTING AND EXEMPLAR - A method of tracking a target includes receiving an observed depth image of the target from a source and analyzing the observed depth image with a prior-trained collection of known poses to find an exemplar pose that represents an observed pose of the target. The method further includes rasterizing a model of the target into a synthesized depth image having a rasterized pose and adjusting the rasterized pose of the model into a model-fitting pose based, at least in part, on differences between the observed depth image and the synthesized depth image. Either the exemplar pose or the model-fitting pose is then selected to represent the target. | 03-10-2011 |
20110069870 | SCREEN SPACE PLANE IDENTIFICATION - A method of finding and defining a plane includes screen-space scanning a plurality of rows of a depth image and interpolating a straight depth line through at least two depth values for each row. A pair of straight boundary lines are then fit to the endpoints of the straight depth lines, and a plane is defined to include these straight boundary lines. | 03-24-2011 |
20110102438 | Systems And Methods For Processing An Image For Target Tracking - An image such as a depth image of a scene may be received, observed, or captured by a device. The image may then be processed. For example, the image may be downsampled, a shadow, noise, and/or a missing potion in the image may be determined, pixels in the image that may be outside a range defined by a capture device associated with the image may be determined, a portion of the image associated with a floor may be detected. Additionally, a target in the image may be determined and scanned. A refined image may then be rendered based on the processed image. The refined image may then be processed to, for example, track a user. | 05-05-2011 |
20110109724 | BODY SCAN - A depth image of a scene may be received, observed, or captured by a device. The depth image may then be analyzed to determine whether the depth image includes a human target. For example, the depth image may include one or more targets including a human target and non-human targets. Each of the targets may be flood filled and compared to a pattern to determine whether the target may be a human target. If one or more of the targets in the depth image includes a human target, the human target may be scanned. A skeletal model of the human target may then be generated based on the scan. | 05-12-2011 |
20120159290 | VALIDATION ANALYSIS OF HUMAN TARGET - Technology for testing a target recognition, analysis, and tracking system is provided. A searchable repository of recorded and synthesized depth clips and associated ground truth tracking data is provided. Data in the repository is used by one or more processing devices each including at least one instance of a target recognition, analysis, and tracking pipeline to analyze performance of the tracking pipeline. An analysis engine provides at least a subset of the searchable set responsive to a request to test the pipeline and receives tracking data output from the pipeline on the at least subset of the searchable set. A report generator outputs an analysis of the tracking data relative to the ground truth in the at least subset to provide an output of the error relative to the ground truth. | 06-21-2012 |
20120163669 | Systems and Methods for Detecting a Tilt Angle from a Depth Image - A depth image of a scene may be received, observed, or captured by a device. A human target in the depth image may then be scanned for one or more body parts such as shoulders, hips, knees, or the like. A tilt angle may then be calculated based on the body parts. For example, a first portion of pixels associated with an upper body part such as the shoulders and a second portion of pixels associated with a lower body part such as a midpoint between the hips and knees may be selected. The tilt angle may then be calculated using the first and second portions of pixels. | 06-28-2012 |
20120287038 | Body Scan - A depth image of a scene may be received, observed, or captured by a device. The depth image may then be analyzed to determine whether the depth image includes a human target. For example, the depth image may include one or more targets including a human target and non-human targets. Each of the targets may be flood filled and compared to a pattern to determine whether the target may be a human target. If one or more of the targets in the depth image includes a human target, the human target may be scanned. A skeletal model of the human target may then be generated based on the scan. | 11-15-2012 |
20130077820 | MACHINE LEARNING GESTURE DETECTION - A virtual skeleton includes a plurality of joints and provides a machine readable representation of a human subject observed with a sensor such as a depth camera. A gesture detection module is trained via machine learning to identify one or more features of a virtual skeleton and indicate if the feature(s) collectively indicate a particular gesture. | 03-28-2013 |
20130101207 | Systems and Methods for Detecting a Tilt Angle from a Depth Image - A depth image of a scene may be received, observed, or captured by a device. A human target in the depth image may then be scanned for one or more body parts such as shoulders, hips, knees, or the like. A tilt angle may then be calculated based on the body parts. For example, a first portion of pixels associated with an upper body part such as the shoulders and a second portion of pixels associated with a lower body part such as a midpoint between the hips and knees may be selected. The tilt angle may then be calculated using the first and second portions of pixels. | 04-25-2013 |
20130129169 | Body scan - A depth image of a scene may be received, observed, or captured by a device. The depth image may then be analyzed to determine whether the depth image includes a human target. For example, the depth image may include one or more targets including a human target and non-human targets. Each of the targets may be flood filled and compared to a pattern to determine whether the target may be a human target. If one or more of the targets in the depth image includes a human target, the human target may be scanned. A skeletal model of the human target may then be generated based on the scan. | 05-23-2013 |
20130129227 | Environment and/or Target Segmentation - A depth image of a scene may be observed or captured by a capture device. The depth image may include a human target and an environment. One or more pixels of the depth image may be analyzed to determine whether the pixels in the depth image are associated with the environment of the depth image. The one or more pixels associated with the environment may then be discarded to isolate the human target and the depth image with the isolated human target may be processed. | 05-23-2013 |
20130251204 | VALIDATION ANALYSIS OF HUMAN TARGET - Technology for testing a target recognition, analysis, and tracking system is provided. A searchable repository of recorded and synthesized depth clips and associated ground truth tracking data is provided. Data in the repository is used by one or more processing devices each including at least one instance of a target recognition, analysis, and tracking pipeline to analyze performance of the tracking pipeline. An analysis engine provides at least a subset of the searchable set responsive to a request to test the pipeline and receives tracking data output from the pipeline on the at least subset of the searchable set. A report generator outputs an analysis of the tracking data relative to the ground truth in the at least subset to provide an output of the error relative to the ground truth. | 09-26-2013 |
20140043438 | Systems and Methods for Detecting a Tilt Angle from a Depth Image - A depth image of a scene may be received, observed, or captured by a device. A human target in the depth image may then be scanned for one or more body parts such as shoulders, hips, knees, or the like. A tilt angle may then be calculated based on the body parts. For example, a first portion of pixels associated with an upper body part such as the shoulders and a second portion of pixels associated with a lower body part such as a midpoint between the hips and knees may be selected. The tilt angle may then be calculated using the first and second portions of pixels. | 02-13-2014 |
20140179421 | CLIENT RENDERING OF LATENCY SENSITIVE GAME FEATURES - Embodiments of the present invention split game processing and rendering between a client and a game server. A rendered video game image is received from a game server and combined with a rendered image generated by the game client to form a single video game image that is presented to a user. Game play may be controlled using a rich sensory input, such as three-dimensional image data and audio data. The three-dimensional image data describes the shape, size and orientation of objects present in a play space. The rich sensory input is communicated to a game server, potentially with some preprocessing, and is also consumed locally on the client, at least in part. In one embodiment, latency sensitive features are the only features processed on the client and rendered on the client. | 06-26-2014 |
20140179436 | CLIENT SIDE PROCESSING OF GAME CONTROLLER INPUT - Embodiments of the present invention enable rich control input data to control video games that are remotely executed. Rich control input includes three-dimensional image data, color video, audio, device orientation data, and touch input. A remotely-executed video game is one executed on a server or other computing device that is networked to a client device receiving the rich control input. Rich control input includes more data than can be uploaded to a game server without degrading game performance. Embodiments of the present invention preprocess the rich control data on the client and into data that may be uploaded to the game server. The rich input stream may be processed in a general way or in a game-specific way. | 06-26-2014 |