Action Recognition On Ava V2 2

評価指標

mAP

評価結果

このベンチマークにおける各モデルのパフォーマンス結果

比較表

モデル名	mAP
videomae-masked-autoencoders-are-data-1	36.1
videomae-masked-autoencoders-are-data-1	26.7
masked-feature-prediction-for-self-supervised	39.8
improved-multiscale-vision-transformers-for	34.4
videomae-masked-autoencoders-are-data-1	39.5
masked-video-distillation-rethinking-masked	31.1
multiscale-vision-transformers	27.3
internvideo-general-video-foundation-models	41.01
masked-video-distillation-rethinking-masked	38.7
videomae-masked-autoencoders-are-data-1	34.3
videomae-masked-autoencoders-are-data-1	39.3
masked-video-distillation-rethinking-masked	34.2
slowfast-networks-for-video-recognition	21.9
hiera-a-hierarchical-vision-transformer	43.3
asymmetric-masked-distillation-for-pre	33.5
holistic-interaction-transformer-network-for	32.6
slowfast-networks-for-video-recognition	23.8
end-to-end-spatio-temporal-action	41.7
masked-video-distillation-rethinking-masked	37.7
slowfast-networks-for-video-recognition	27.5
videomae-v2-scaling-video-masked-autoencoders	42.6
videomae-masked-autoencoders-are-data-1	37.8
videomae-masked-autoencoders-are-data-1	31.8
multiscale-vision-transformers	26.1
actor-context-actor-relation-network-for	31.72
towards-long-form-video-understanding-1	31.0
unmasked-teacher-towards-training-efficient	39.8
multiscale-vision-transformers	28.7
on-the-benefits-of-3d-pose-and-tracking-for	45.1
videomae-masked-autoencoders-are-data-1	36.5
slowfast-networks-for-video-recognition	27.1
multiscale-vision-transformers	26.8
memvit-memory-augmented-multiscale-vision	35.4
multiscale-vision-transformers	24.5
masked-video-distillation-rethinking-masked	41.1
masked-video-distillation-rethinking-masked	40.1
object-region-video-transformers-1	26.6
multiscale-vision-transformers	27.5