List your company

Spatiotemporal Action Localization on AVA 2.2

41.7mAPSTAR/L*
22.25227.30132.3537.399Dec 6, 2022Dec 29, 2022Jan 21, 2023Feb 13, 2023Mar 8, 2023Mar 31, 2023Apr 24, 2023

Evaluation results

MethodLinks
STAR/L*Pretraining=CLIP→K700, Views=1, Resolution=320, Backbone=ViViT/L, End-to-end=true, Trained on=AVA-Kinetics2023.0441.7
InternVideoHead=Linear2022.1241.01
InternVideo*Pretraining=7 different datasets, Backbone=Uniformer v2, End-to-end=false, Trained on=AVA-Kinetics2023.0441
VideoMAEPretraining=SSL K700 → Sup. K700, Resolution=256, Backbone=ViViT/L, End-to-end=false2023.0439.3
STAR/LPretraining=CLIP→K700, Views=1, Resolution=320, Backbone=ViViT/L, End-to-end=true2023.0439.2
STAR/LPretraining=JFT→WTS, Views=1, Resolution=320, Backbone=ViViT/L, End-to-end=true2023.0439
MaskFeatHead=not specified2022.1238.8
STAR/BPretraining=JFT→WTS, Views=1, Resolution=320, Backbone=ViViT/B, End-to-end=true2023.0436.3
Co-finetuningPretraining=JFT, WTS→K700, MiT, SSv2, Views=1, Resolution=320, Backbone=ViViT/L, End-to-end=false2023.0436.1
MeMVITPretraining=K700, Resolution=312, Backbone=MViT v2, End-to-end=false2023.0434.4
STAR/BPretraining=CLIP→K700, Views=1, Resolution=320, Backbone=ViViT/B, End-to-end=true2023.0433.9
TubeRPretraining=Instagram65M [36]→K400, Views=2, Resolution=256, Backbone=CSN-152, End-to-end=true2023.0433.6
ACARHead=ACAR [77], Ensemble=true2022.1233.3
ACARPretraining=K700, Views=6, Resolution=320, Backbone=SlowFast, End-to-end=false2023.0433.3
Co-finetuningPretraining=IN21K→K700, MiT, SSv2, Views=1, Resolution=320, Backbone=ViViT/L, End-to-end=false2023.0432.8
AIAPretraining=K700, Views=18, Resolution=320, Backbone=SlowFast, End-to-end=false2023.0432.3
STAR/BPretraining=IN21K→K400, Views=1, Resolution=320, Backbone=ViViT/B, End-to-end=true2023.0430
WOOPretraining=K600, Views=1, Resolution=320, Backbone=SlowFast, End-to-end=true2023.0428.3
UnifiedPretraining=K400, Views=6, Resolution=320, Backbone=SlowFast, End-to-end=false2023.0427.7
MViT-BPretraining=K400, Views=1, Backbone=MViT, End-to-end=false2023.0427.3
Action TransformerPretraining=K400, Views=1, Resolution=400, Backbone=I3D, End-to-end=true2023.0423