VideoChat’s IT

• Paper: VideoChat: Chat-Centric Video Understanding
• GitHub Link
• Publisher: Arxiv
• Author Affiliation: Shanghai AI Laboratory & Nanjing University & The University of Hong Kong & Chinese Academy of Sciences
• Type
  • SFT
  • RLHF
• Multi-turn
  • ✔
  • ✖
• Input Modalities $\rightarrow$ Output Modalities
  (I: Image, V: Video, A: Audio, 3D: Point Cloud, T: Text, B: Bounding box, Tab: Table, Web: Web page)
  • V+T $\rightarrow$ T
• Source
  • WebVid
• Method
  • Auto.
• I/V/A Scale
  • I
    • Not report
  • V
    • 8K
  • A
    • Not report
• Dialog Turn
  • 1.82
• Instance Scale
  • 11K