Video motion processing including static scene determination, occlusion detection, frame rate conversion, and adjusting compression ratio

A system comprising:

• one or more depth cameras to determine a depth of a moving object;
• a processor;
• memory including a set of instructions which, when executed by the processor, cause the processor to receive a motion vector for a frame of a video;
• and
• logic implemented at least partly in one or more of configurable logic or fixed-functionality hardware logic, the logic to:
• increase a frame rate for the video;
• detect a motion of an object within one or more frames of the video to determine whether a scene in the video is static or changing;
• detect a depth of the object based on input from the one or more depth cameras;
• when it is determined that the scene is changing and the object is moving, estimate an occlusion of the object in an intermediate frame with another object, wherein the estimated occlusion of the object is determined based at least in part on the depth of the object;
• and
• interpolate the intermediate frame in a frame rate conversion process, wherein the estimated occlusion is used to determine where to interpolate the object with respect to one or more other objects.

A graphics processing apparatus comprising logic implemented at least partly in one or more of configurable logic or fixed-functionality hardware logic, the logic to:

• increase a frame rate for a video;
• detect a motion of an object within one or more frames of the video to determine whether a scene in the video is static or changing;
• detect a depth of the object based on input from one or more depth cameras;
• when it is determined that the scene is changing and the object is moving, estimate an occlusion of the object in an intermediate frame with another object, wherein the estimated occlusion of the object is determined based at least in part on the depth of the object;
• and
• interpolate the intermediate frame in a frame rate conversion process, wherein the estimated occlusion is used to determine where to interpolate the object with respect to one or more other objects.

At least one non-transitory computer readable storage medium comprising a set of instructions which, when executed by a computing device, cause the computing device to:

• increase a frame rate for a video;
• detect a motion of an object within one or more frames of the video to determine whether a scene in the video is static or changing;
• detect a depth of the object based on input from one or more depth cameras;
• when it is determined that the scene is changing and the object is moving, estimate an occlusion of the object in an intermediate frame with another object, wherein the estimated occlusion of the object is determined based at least in part on the depth of the object;
• and
• interpolate the intermediate frame in a frame rate conversion process, wherein the estimated occlusion is used to determine where to interpolate the object with respect to one or more other objects.

A method comprising:

• increasing a frame rate for a video;
• detecting a motion of an object within one or more frames of the video to determine whether a scene in the video is static or changing;
• detecting a depth of the object based on input from one or more depth cameras;
• when it is determined that the scene is changing and the object is moving, estimating an occlusion of the object in an intermediate frame with another object, wherein the estimated occlusion of the object is determined based at least in part on the depth of the object;
• and
• interpolating the intermediate frame in a frame rate conversion process, wherein the estimated occlusion is used to determine where to interpolate the object with respect to one or more other objects.