Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Egocentric Video Understanding on Ego-in-Exo PerceptionMCQ (test)
Loading...
83.1
Action Understanding
VisCoP + DnA
74.572
76.786
79
81.214
Jun 25, 2026
Action Understanding
Task Regions
Human-Object Interaction (HOI)
Hand Identification
Average Performance
Updated 29d ago
Evaluation Results
Method
Method
Links
Action Understanding
Task Regions
Human-Object Interaction (HOI)
Hand Identification
Average Performance
VisCoP + DnA
Base model=VisCoP, Att...
2026.06
83.1
87.1
79.2
65.1
78.6
VisCoP
2026.06
81.8
86.1
79.3
65.1
78.1
VisCoP + Diff. Attn.
Base model=VisCoP, Att...
2026.06
78.8
83
77.5
64.9
76.1
VideoLLaMA3
Model Type=Video LLM
2026.06
74.9
75.8
75.2
65.3
72.8
Feedback
Search any
task
Search any
task