Action Recognition In Videos on UCF101

Metric: 3-fold Accuracy (higher is better)

LeaderboardDataset

Loading chart...

Results

Submit a result

Sort:

#	Model↕	3-fold Accuracy▼	Extra Data	Paper	Date↕	Code
1	STM (ImageNet+Kinetics pretrain)	96.2	No	STM: SpatioTemporal and Motion Encoding for Acti...	2019-08-07	-
2	3D-SqueezeNet	74.94	No	Resource Efficient 3D Convolutional Neural Netwo...	2019-04-04	Code
3	3D-ShuffleNetV2 0.25x	56.52	No	Resource Efficient 3D Convolutional Neural Netwo...	2019-04-04	Code
4	3D-MobileNetV2 0.2x	55.56	No	Resource Efficient 3D Convolutional Neural Netwo...	2019-04-04	Code
5	Baseline UCF101	43.9	No	UCF101: A Dataset of 101 Human Actions Classes F...	2012-12-03	Code

#1STM (ImageNet+Kinetics pretrain)SOTA
96.2
3-fold Accuracy· 2019-08-07
STM: SpatioTemporal and Motion Encoding for Action Recognition
#23D-SqueezeNetSOTA
74.94
3-fold Accuracy· 2019-04-04
Resource Efficient 3D Convolutional Neural Networks Code
#33D-ShuffleNetV2 0.25x
56.52
3-fold Accuracy· 2019-04-04
Resource Efficient 3D Convolutional Neural Networks Code
#43D-MobileNetV2 0.2x
55.56
3-fold Accuracy· 2019-04-04
Resource Efficient 3D Convolutional Neural Networks Code
#5Baseline UCF101SOTA
43.9
3-fold Accuracy· 2012-12-03
UCF101: A Dataset of 101 Human Actions Classes From Videos in The Wild Code