SJTU, CUHK, Shanghai AI Lab · Video Understanding
SeC - Concept-Driven Video Object Segmentation
SeC shifts video object segmentation from appearance matching to progressive concept construction using Large Vision-Language Models, achieving 11.8-point improvement over SAM 2.1. Introduces SeCVOS benchmark with 160 annotated videos for evaluating semantic understanding under complex visual variations and occlusions.
Open the source ↗Similar Video & Animation releases
- OverSeer
University of Oxford and Meta (Runjia Li, Andrea Vedaldi, Shikun Liu et al.) · Multi-agent world model · Week 45, 2026
- UniAction
Bowen Xue, Zizhang Li, Hong-Xing Yu, Jiajun Wu · Unified action world model · Week 45, 2026
- WorldCrafter
ARC Lab (Tencent IEG) & Peking University · Explorable World Model · Week 40, 2026
- Odyssey-3
Odyssey · World Model · Week 39, 2026
- LingBot-World 2.0
Robbyant · Interactive World Model · Week 37, 2026
- SolarWM
CUHK-Shenzhen, NUS, HKUST, NVIDIA and MSRA · Long-Horizon World Model · Week 36, 2026
Published by Quanta AI Labs, an AI agency in India. Browse every week in AI News Weekly.