Learning Optimal Lattice Vector Quantizers for End-to-end Neural Image Compression
This work addresses the problem of inefficient quantization in neural image compression for applications requiring high compression quality, representing an incremental advance by adapting LVQ to non-uniform distributions.
The paper tackles the suboptimality of traditional lattice vector quantization (LVQ) for neural image compression by proposing a learning method to design rate-distortion optimal LVQ codebooks based on latent feature statistics, resulting in significant improvements in rate-distortion performance while maintaining computational efficiency similar to uniform scalar quantization.
It is customary to deploy uniform scalar quantization in the end-to-end optimized Neural image compression methods, instead of more powerful vector quantization, due to the high complexity of the latter. Lattice vector quantization (LVQ), on the other hand, presents a compelling alternative, which can exploit inter-feature dependencies more effectively while keeping computational efficiency almost the same as scalar quantization. However, traditional LVQ structures are designed/optimized for uniform source distributions, hence nonadaptive and suboptimal for real source distributions of latent code space for Neural image compression tasks. In this paper, we propose a novel learning method to overcome this weakness by designing the rate-distortion optimal lattice vector quantization (OLVQ) codebooks with respect to the sample statistics of the latent features to be compressed. By being able to better fit the LVQ structures to any given latent sample distribution, the proposed OLVQ method improves the rate-distortion performances of the existing quantization schemes in neural image compression significantly, while retaining the amenability of uniform scalar quantization.