Login / Signup

Elysium: Exploring Object-level Perception in Videos via MLLM.

Han WangYanjie WangYongjie YeYuxiang NieCan Huang
Published in: CoRR (2024)
Keyphrases
  • object level
  • meta level
  • pixel level
  • low level
  • high level
  • higher level
  • object class
  • video sequences
  • video frames
  • object recognition
  • video data
  • multiscale
  • viewpoint
  • superpixels
  • moving camera