CVRL (R3D-152 2x; K600 pretrain)

Reported on 2 benchmarks across 2 tasks · 1 paper · 2 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Robots1 result

Activity RecognitiononKinetics-400
Top-1 accuracy %· uses extra data· 2020-08-09
71.6
best: 77.6 (XKD (ViT-B/112/16))
SOTA
Spatiotemporal Contrastive Video Representation Learning arXiv:2008.03800

Time Series1 result

Action RecognitiononKinetics-400
Top-1 accuracy %· uses extra data· 2020-08-09
71.6
best: 77.6 (XKD (ViT-B/112/16))
SOTA
Spatiotemporal Contrastive Video Representation Learning arXiv:2008.03800