Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multimodal Action Recognition on Kinetic-Sounds ViT
Loading...
74.2
Accuracy
DMIL
67.3048
69.0949
70.885
72.6751
Jun 10, 2026
Accuracy
mAP
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
mAP
DMIL
Backbone=ViT
2026.06
74.2
81.83
Ensemble
Backbone=ViT
2026.06
71.8
79.03
MLB
Backbone=ViT
2026.06
71.22
77.29
MCR
Backbone=ViT
2026.06
69.51
76.33
MMML
Backbone=ViT
2026.06
69.14
75.82
I2M2
Backbone=ViT
2026.06
68.93
73.55
FCL
Backbone=ViT
2026.06
68.9
75.38
AGM
Backbone=ViT
2026.06
68.51
75.43
QMF
Backbone=ViT
2026.06
68.1
72.9
OGM
Backbone=ViT
2026.06
68.02
73.42
PMR
Backbone=ViT
2026.06
67.99
73.36
Joint
Backbone=ViT
2026.06
67.81
73.11
MMIB
Backbone=ViT
2026.06
67.57
73.52
Feedback
Search any
task
Search any
task