CodecSight: Leveraging Video Codec Signals for Efficient Streaming VLM Inference
Published in arXiv, 2026
Recommended citation: Yulin Zou, Yan Chen, Wenyan Chen, JooYoung Park, Shivaraman Nitin, Luo Tao, Francisco Romero, Dmitrii Ustiugov. (2026). "CodecSight: Leveraging Video Codec Signals for Efficient Streaming VLM Inference." arXiv:2604.06036. https://arxiv.org/abs/2604.06036
CodecSight is a codec-guided streaming video analytics system for efficient vision-language model inference.
| Paper | arXiv |
Recommended citation: Yulin Zou, Yan Chen, Wenyan Chen, JooYoung Park, Shivaraman Nitin, Luo Tao, Francisco Romero, Dmitrii Ustiugov. (2026). “CodecSight: Leveraging Video Codec Signals for Efficient Streaming VLM Inference.” arXiv:2604.06036.