NTIRE 2026挑战：比特流损坏视频恢复的方法与结果
计算机视觉与模式识别
2026-04-15 v3
摘要
本文报告了NTIRE 2026挑战中关于比特流损坏视频恢复（BSCVR）的工作。该挑战旨在推动研究从受损比特流中恢复视觉连贯视频，其解码常产生严重的时空失真和内容扭曲。基于最近在比特流损坏视频恢复方面的进展，挑战提供了在真实损坏设置下评估恢复方法的通用基准。我们描述了数据集、评估协议和参赛方法，并总结了最终结果及主要技术趋势。该挑战凸显了这一新兴任务的难度，为未来在实际比特流损坏条件下进行鲁棒视频恢复提供了有益见解。
引用
@article{arxiv.2604.06945,
title = {NTIRE 2026 Challenge on Bitstream-Corrupted Video Restoration: Methods and Results},
author = {Wenbin Zou and Tianyi Liu and Kejun Wu and Huiping Zhuang and Zongwei Wu and Zhuyun Zhou and Radu Timofte and Kim-Hui Yap and Lap-Pui Chau and Yi Wang and Shiqi Zhou and Xiaodi Shi and Yuxiang Chen and Yilian Zhong and Shibo Yin and Yushun Fang and Xilei Zhu and Yahui Wang and Chen Lu and Zhitao Wang and Lifa Ha and Hengyu Man and Xiaopeng Fan and Priyansh Singh and Sidharth and Krrish Dev and Soham Kakkar and Vinit Jakhetiya and Ovais Iqbal Shah and Wei Zhou and Linfeng Li and Qi Xu and Zhenyang Liu and Kepeng Xu and Tong Qiao and Jiachen Tu and Guoyi Xu and Yaoxin Jiang and Jiajia Liu and Yaokun Shi},
journal= {arXiv preprint arXiv:2604.06945},
year = {2026}
}
备注
15 pages, 8 figures, 1 table, CVPRW2026 NTIRE Challenge Report