CPC G06V 10/7715 (2022.01) [H04N 19/91 (2014.11)] | 5 Claims |
1. A method of compressing feature data, the method comprising:
receiving C×W/N×H/N feature data, wherein C is a number of channels, W is a width and H is a height of input data corresponding to the feature data, and N is a downsampling factor;
performing residual encoding on the received feature data to generate enhanced feature data by adding refinement values to the received feature data;
applying a two-dimensional convolution operation on the enhanced feature data to generate reduced feature data, wherein the received feature data is reduced about the spatial dimensions, W/N and H/N and the channel dimension, C;
generating a heatmap based on the reduced feature data, wherein the heat map is an array of scaling factors to be applied at each spatial location;
scaling the reduced feature data using the generated heatmap; and
entropy encoding the scaled reduced feature data to generate a bitstream.
|