Neural Image Compression Using Masked Sparse Visual Representation

TL;DR

This paper introduces M-AdaCode, a neural image compression method using masked sparse visual representations that adaptively balances bitrate and reconstruction quality through learned codebooks and masking strategies.

Contribution

The paper proposes a novel Masked Adaptive Codebook learning method that enables flexible rate-distortion tradeoffs in neural image compression by masking and combining semantic-dependent codebooks.

Findings

Effective rate-distortion tradeoff achieved

High-quality reconstruction with adaptive codebook weighting

Demonstrated improvements on JPEG-AI dataset

Abstract

We study neural image compression based on the Sparse Visual Representation (SVR), where images are embedded into a discrete latent space spanned by learned visual codebooks. By sharing codebooks with the decoder, the encoder transfers integer codeword indices that are efficient and cross-platform robust, and the decoder retrieves the embedded latent feature using the indices for reconstruction. Previous SVR-based compression lacks effective mechanism for rate-distortion tradeoffs, where one can only pursue either high reconstruction quality or low transmission bitrate. We propose a Masked Adaptive Codebook learning (M-AdaCode) method that applies masks to the latent feature subspace to balance bitrate and reconstruction quality. A set of semantic-class-dependent basis codebooks are learned, which are weighted combined to generate a rich latent feature for high-quality reconstruction. The combining weights are adaptively derived from each input image, providing fidelity information with additional transmission costs. By masking out unimportant weights in the encoder and recovering them in the decoder, we can trade off reconstruction quality for transmission bits, and the masking rate controls the balance between bitrate and distortion. Experiments over the standard JPEG-AI dataset demonstrate the effectiveness of our M-AdaCode approach.