NVIDIA Research Taiwan
NVIDIA Research Taiwan
Home
News
Members
Research
Publications
Contact
Light
Dark
Automatic
Video Anomaly Understanding
VADER: Towards Causal Video Anomaly Understanding with Relation-Aware Large Language Models
Video anomaly understanding (VAU) aims to provide detailed interpretation and semantic comprehension of anomalous events within videos, addressing limitations of traditional methods that focus solely on detecting and localizing anomalies. However, …
Cite
×