Exploiting Latent Properties to Optimize Neural Codecs
- 1 January 2025
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 34 (10577149) , 306-319
Abstract
End-to-end image and video codecs are becoming increasingly competitive, compared to traditional compression techniques that have been developed through decades of manual engineering efforts. These trainable codecs have many advantages over traditional techniques, such as their straightforward adaptation to perceptual distortion metrics and high performance in specific fields thanks to their learning ability. However, current state-of-the-art neural codecs do not fully exploit the benefits of vector quantization and the existence of the entropy gradient in decoding devices. In this paper, we propose to leverage these two properties (vector quantization and entropy gradient) to improve the performance of off-the-shelf codecs. Firstly, we demonstrate that using non-uniform scalar quantization cannot improve performance over uniform quantization. We thus suggest using predefined optimal uniform vector quantization to improve performance. Secondly, we show that the entropy gradient, available at the decoder, is correlated with the reconstruction error gradient, which is not available at the decoder. We therefore use the former as a proxy to enhance compression performance. Our experimental results show that these approaches save between 1 to 3% of the rate for the same quality across various pre-trained methods. In addition, the entropy gradient based solution improves traditional codec performance significantly as well.Keywords
All Related Versions
This publication has 35 references indexed in Scilit:
- Joint Global and Local Hierarchical Priors for Learned Image Compression2022
- Checkerboard Context Model for Efficient Learned Image Compression2021
- Content Adaptive and Error Propagation Aware Deep Video Compression2020
- Scale-Space Flow for End-to-End Optimized Video Compression2020
- Learned Image Compression With Discretized Gaussian Mixture Likelihoods and Attention Modules2020
- DVC: An End-To-End Deep Video Compression Framework2019
- Multiple-gradient descent algorithm (MGDA) for multiobjective optimizationComptes Rendus Mathématique, 2012
- Vector Quantization and Signal Compression1992
- Multilayer feedforward networks are universal approximatorsNeural Networks, 1989
- Spectra of Quantized SignalsBell System Technical Journal, 1948