VGG19 + 393K webcam images

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 RecognitiononActivityNet
mAP· uses extra data· 2015-12-22
53.8
best: 96.9 (Text4Vis (w/ ViT-L))
SOTA
Do Less and Achieve More: Training CNNs for Action Recognition Utilizing Action Images from the Web arXiv:1512.07155

Time Series1 result

Action RecognitiononActivityNet
mAP· uses extra data· 2015-12-22
53.8
best: 96.9 (Text4Vis (w/ ViT-L))
SOTA
Do Less and Achieve More: Training CNNs for Action Recognition Utilizing Action Images from the Web arXiv:1512.07155