R(2+1)D-18 (DistInit pretraining)

Reported on 4 benchmarks across 2 tasks · 1 paper

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

Robots2 results

Activity RecognitiononHMDB-51
Average accuracy of 3 splits· uses extra data· 2019-01-26
54.8
best: 88.7 (VideoMAE V2-g)
DistInit: Learning Video Representations Without a Single Labeled Video arXiv:1901.09244
Activity RecognitiononUCF101
3-fold Accuracy· 2019-01-26
85.8
best: 99.7 (FTP-UniFormerV2-L/14)
DistInit: Learning Video Representations Without a Single Labeled Video arXiv:1901.09244

Time Series2 results

Action RecognitiononHMDB-51
Average accuracy of 3 splits· uses extra data· 2019-01-26
54.8
best: 88.7 (VideoMAE V2-g)
DistInit: Learning Video Representations Without a Single Labeled Video arXiv:1901.09244
Action RecognitiononUCF101
3-fold Accuracy· 2019-01-26
85.8
best: 99.7 (FTP-UniFormerV2-L/14)
DistInit: Learning Video Representations Without a Single Labeled Video arXiv:1901.09244