A decoupled spatial-channel inverted bottleneck for image compression

Residual block has achieved great success in deep networks to eliminate accuracy degradation, and there emerges a large number of variants with more competitive performance. However, these blocks are introduced for high-level tasks that only encode the input image into semantic and structural features but do not need to reconstruct. So for the low-level task like image compression where reconstruction quality contributes significantly to the rate-distortion performance, the structure of the residual block needs modification for more suitable implementation. In this paper, we revisit the existing residual blocks and discover two key principles summarized as two decouplings: spatial-channel decoupling and linear-nonlinear decoupling. We propose an efficient nonlinear transform based on the principles dubbed decoupled spatial-channel inverted bottleneck(DSCIB), which has a linear-spatial branch for rough reconstruction and a nonlinear-channel branch to provide detailed featrues. We employ the DSCIB module in the joint autoregression model to build an overall network. Experimental results show that our method achieves comparable performance with the existing learning-based image compression methods at high bitrate while reducing 38% FLOPs.