A technology can be provided for change invariant scene recognition of an environment by an agent that uses image sensors to capture images. An example method can include grouping the images into situation image sets to form an image database. A navigation image can be captured by the agent at the agent's current location. Another operation can be searching through images by separately loaded situation image set to identify images that are a probable match for the navigation image. Error values can be computed between the navigation image and situation image sets. A probability can be calculated that situation image sets are to be used as a relevant situation image set based on the error values. Then the relevant situation image set can be searched for matching purposes with the navigation image while the error value is below a defined threshold.

Change invariant scene recognition by an agent

Systems, methods, and computer media for generating an avatar reflecting a player's current appearance. Data describing the player's current appearance is received. The data includes a visible spectrum image of the player, a depth image including both the player and a current background, and skeletal data for the player. The skeletal data indicates an outline of the player's skeleton. Based at least in part on the received data, one or more of the following are captured: a facial appearance of the player; a hair appearance of the player; a clothing appearance of the player; and a skin color of the player. A 3D avatar resembling the player is generated by combining the captured facial appearance, hair appearance, clothing appearance, and/or skin color with predetermined avatar features.

Generation of avatar reflecting player appearance

Systems and techniques are provided for tracking puts and takes of inventory items by subjects in an area of real space. A plurality of cameras with overlapping fields of view produce respective sequences of images of corresponding fields of view in the real space. In one embodiment, the system includes first image processors, including subject image recognition engines, receiving corresponding sequences of images from the plurality of cameras. The first image processors process images to identify subjects represented in the images in the corresponding sequences of images. The system includes second image processors, including background image recognition engines, receiving corresponding sequences of images from the plurality of cameras. The second image processors mask the identified subjects to generate masked images. Following this, the second image processors process the masked images to identify and classify background changes represented in the images in the corresponding sequences of images.

Predicting inventory events using semantic diffing

Systems, methods, apparatuses, and computer readable media are disclosed for improving, in some examples, real time location systems with multiple location technologies. In one embodiment, a method is provided including receiving blink data from a location tag associated with a first sensor; receiving proximity data generated based on communications between the first sensor and a second sensor, the proximity data including a sensor identifier; calculating location data associated with the location tag based on the blink data; and determining sensor position calculation data associated with the first sensor based on the proximity data.

Method, apparatus, and computer program product improving real time location systems with multiple location technologies

Systems and techniques are provided for tracking multi joint subjects in real space having a plurality of cameras. The field of view of each camera overlaps with at least one other camera. The plurality of cameras produce respective sequences of images of corresponding fields of view in the real space. A processing system is coupled to the plurality of cameras. In one embodiment, the processing system comprises image recognition engines receiving sequence of images from the plurality of cameras and generating corresponding arrays of joint data structures. A tracking engine is configured to receive the arrays of joint data structures and generate candidate joints having coordinates in the real space. The processing system includes the logic to identify sets of candidate joints having coordinates in the real space as multi-joint subjects in the real space.

Subject identification and tracking using image recognition

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.

Michael Taylor Janney Mt

Papers

Validation analysis of human target