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

2026-06-11

AI Summary

Existing learned video compression methods suffer from transform-induced distortion in high-quality regions due to their reliance on irreversible transforms, limiting fidelity in reconstruction. This work proposes InnVC, the first framework within a unified architecture capable of supporting a broad operating range from low bitrates to high-fidelity reconstruction. The key innovation lies in integrating an invertible primary transform with implicit conditional modeling: by preserving an invertible path prior to quantization, it decouples strongly correlated content from hard-to-model details and introduces a content-adaptive context injection mechanism. Additionally, a scheduled masking strategy is designed to enhance entropy coding efficiency. Experiments demonstrate that InnVC significantly outperforms x265 on the UVG and MCL-JCV datasets, achieving BD-rate reductions of 21.66% in PSNR and 46.06% in MS-SSIM on UVG, while supporting high-quality compression across a PSNR range exceeding 20 dB.

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 reconstruction quality subject to both quantization and transform approximation errors. This limitation becomes particularly restrictive at higher quality points, where quantization errors are small and transform-induced distortion dominates. To address this, we propose InnVC, an Invertible neural network based Video Codec for wide-range and high-fidelity compression. The core idea is to preserve an invertible main transform path prior to quantization, while injecting content-adaptive context through a compact implicit conditioning field. This decouples strongly correlated video content from harder-to-model fine details, allowing different components to specialize in complementary reconstruction tasks for more efficient compression. To further improve compressibility, we introduce a scheduled masking strategy that progressively concentrates informative content into fewer latent channels for more effective entropy coding. Experiments on the UVG and MCL-JCV benchmarks show that InnVC achieves strong compression performance over a broad quality range, being particularly effective in the high-quality regime, yielding BD-rate reductions of 21.66% in PSNR and 46.06% in MS-SSIM relative to x265 on UVG. To the best of our knowledge, InnVC is the first neural video codec covers operating poins from low bitrate to high fidelity within a single architecture scale, spanning more than 20 dB in PSNR.

Problem

• video compression
• invertible transform
• high-fidelity
• rate-distortion
• neural codec

Innovation

• Invertible Neural Transform
• Implicit Conditioning
• High-Fidelity Video Compression
• Scheduled Masking
• Neural Video Codec