Abstract: The present disclosure may be a deep learning-based method for improving feature map compression efficiency, wherein feature maps can be selectively transmitted from an encoder, and untransmitted feature maps can be predicted and generated after restoring the image quality of the transmitted feature maps using a deep neural network.

Abstract: Disclosed herein are a method, an apparatus and a storage medium for encoding/decoding using a transform-based feature map. An optimal basis vector is extracted from one or more feature maps, and a transform coefficient is acquired through a transform using the basis vector. The basis vector and the transform coefficient may be transmitted through a bitstream. In an embodiment, one or more feature maps are reconstructed using the basis vector and the transform coefficient, which are decoded from the bitstream.

Abstract: Disclosed herein are a method, an apparatus, and a storage medium for encoding/decoding a multi-resolution feature map. In an aspect, there is provided an encoding method, including extracting feature maps from an input image, performing packing on the extracted feature maps, and performing encoding on the packed feature maps, wherein the feature maps include multiple feature maps having different resolutions. In another aspect, there is provided a decoding method, including generating packed feature maps by decoding an input bitstream, and generating a multi-resolution feature map by performing unpacking on the packed feature maps, wherein the multi-resolution feature map includes multiple feature maps having different resolutions.