Traditional video codecs optimized for pixel fidelity collapse at ultra-low bitrates and produce severe artifacts. This failure arises from a fundamental misalignment between pixel accuracy and human perception. We propose a semantic video compression framework named DiSCo that transmits only the most meaningful information while relying on generative priors for detail synthesis. The source video is decomposed into three compact modalities: a textual description, a spatiotemporally degraded video, and optional sketches or poses that respectively capture semantic, appearance, and motion cues. A conditional video diffusion model then reconstructs high-quality, temporally coherent videos from these compact representations. Temporal forward filling, token interleaving, and modality-specific codecs are proposed to improve multimodal generation and modality compactness. Experiments show that our method outperforms baseline semantic and traditional codecs by 2-10X on perceptual metrics at low bitrates.
@article{arxiv.2512.00408,
title = {Low-Bitrate Video Compression through Semantic-Conditioned Diffusion},
author = {Lingdong Wang and Guan-Ming Su and Divya Kothandaraman and Tsung-Wei Huang and Mohammad Hajiesmaili and Ramesh K. Sitaraman},
journal= {arXiv preprint arXiv:2512.00408},
year = {2026}
}