NON-LOCAL MEMORY NETWORK FOR SEMI-SUPERVISED VIDEO OBJECT SEGMENTATION

摘要

#$%^&*AU2019213369A120200430.pdf#####Abstract Certain aspects involve using a space-time memory network to locate one or more target objects in video content for segmentation or other object classification. In one example, a video editor generates a query key map and a query value map by applying a space-time memory network to features of a query frame from video content. The video editor retrieves a memory key map and a memory value map that are computed, with the space-time memory network, from a set of memory frames from the video content. The video editor computes memory weights by applying a similarity function to the memory key map and the query key map. The video editor classifies content in the query frame as depicting the target feature using a weighted summation that includes the memory weights applied to memory locations in the memory value map.2/13 200 ,, 202 Access a queryframe Generate a query key map and a Generate memory key maps and 204'e query value map from features of the 205 memory value maps for memory query frame frames 206 Retrieve memory key maps and memory value maps from a memory network Compute memory weights by applying a similarity function to the memory key 208 maps and the query key map Generate a segmentation mask for the query frame from a weighted 210 summation that includes the memory weights applied to the memory value maps FIG. 2