PulseAugur
EN
LIVE 21:16:57

New Invertible Neural Network Video Codec Achieves High-Fidelity Compression

By PulseAugur Editorial · [2 sources] ·

Researchers have developed InnVC, a novel invertible neural network-based video codec designed for high-fidelity compression. Unlike previous methods that suffer from approximation errors, InnVC maintains an invertible transform path before quantization and uses implicit conditioning to handle video content adaptively. This approach allows for more efficient compression, particularly at higher quality settings, and has demonstrated significant BD-rate reductions compared to conventional codecs like x265 on benchmarks such as UVG and MCL-JCV. AI

RANK_REASON The cluster contains a research paper detailing a new technical approach to video compression.

Read on arXiv cs.CV →

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

New Invertible Neural Network Video Codec Achieves High-Fidelity Compression

COVERAGE [2]

  1. arXiv cs.CV TIER_1 English(EN) · Siyue Teng, Ho Man Kwan, Yuxuan Jiang, Fan Zhang, David Bull ·

    High-Fidelity Video Compression based on Invertible Neural Transform and Implicit Conditioning

    arXiv:2606.13957v1 Announce Type: cross Abstract: Learning-based video compression has recently achieved competitive rate-distortion performance compared to conventional video codecs. However, most existing methods rely on non-invertible analysis-synthesis transforms, with recons…

  2. arXiv cs.CV TIER_1 English(EN) · David Bull ·

    High-Fidelity Video Compression based on Invertible Neural Transform and Implicit Conditioning

    Learning-based video compression has recently achieved competitive rate-distortion performance compared to conventional video codecs. However, most existing methods rely on non-invertible analysis-synthesis transforms, with reconstruction quality subject to both quantization and …

RELATED TOPICS