paper-with-me

홈 › Papers

Latent-Shift: Gradient of Entropy Helps Neural Codecs

2023-08-01 · Muhammet Balcilar, Bharath Bhushan Damodaran, Karam Naser, Franck Galpin, Pierre Hellier

End-to-end image/video codecs are getting 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 easy adaptation on perceptual distortion metrics and high performance on specific domains thanks to their learning ability. However, state of the art neural codecs does not take advantage of the existence of gradient of entropy in decoding device. In this paper, we theoretically show that gradient of entropy (available at decoder side) is correlated with the gradient of the reconstruction error (which is not available at decoder side). We then demonstrate experimentally that this gradient can be used on various compression methods, leading to a $1-2\%$ rate savings for the same quality. Our method is orthogonal to other improvements and brings independent rate savings.

📄 PDF Abstract BibTeX arXiv:2308.00725

Code (0)

등록된 구현이 없습니다.

Tasks

Decoder

Similar Papers 제목 키워드 기반

Exploiting Latent Properties to Optimize Neural Codecs

2025-01-02 · Muhammet Balcilar, Bharath Bhushan Damodaran, Karam Naser, Franck Galpin 외

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 …

DecoderQuantization

Hybrid Spatial-Temporal Entropy Modelling for Neural Video Compression

2022-07-13 · Jiahao Li, Bin Li, Yan Lu

For neural video codec, it is critical, yet challenging, to design an efficient entropy model which can accurately predict the probability distribution of the quantized latent representation. However, most existing video…

QuantizationVideo Compression

Channel-wise Autoregressive Entropy Models for Learned Image Compression

2020-07-17 · David Minnen, Saurabh Singh

In learning-based approaches to image compression, codecs are developed by optimizing a computational model to minimize a rate-distortion objective. Currently, the most effective learned image codecs take the form of an …

GPUImage Compression

HyperVQ: Enabling Hyperprior Entropy Modeling for VQ-Based Generative Image Compression

2025-12-08 · Niu Yi, Xu Tianyi, Ma Mingming, Wang Xinkun arxiv

Vector Quantization (VQ) based generative image compression has achieved remarkable perceptual quality. However, existing VQ codecs suffer from two fundamental limitations. First, they lack efficient content-adaptive ent…

Image Compression

Frequency Disentangled Features in Neural Image Compression

2023-08-04 · Ali Zafari, Atefeh Khoshkhahtinat, Piyush Mehta, Mohammad Saeed Ebrahimi Saadabadi 외

The design of a neural image compression network is governed by how well the entropy model matches the true distribution of the latent code. Apart from the model capacity, this ability is indirectly under the effect of h…

DisentanglementImage CompressionQuantization