Temporal Action Localization paper

Paper & code

weakly-supervised

• W-TALC (ECCV18) W-TALC: Weakly-supervised Temporal Activity Localization and Classification tensorflow, pytorch
• STPN (CVPR18) Weakly Supervised Action Localization by Sparse Temporal Pooling Network tensorflow, pytorch
• ASSG (ACM19) Adversarial Seeded Sequence Growing for Weakly-Supervised Temporal Action Localization
• TSM (ICCV19) Temporal Structure Mining for Weakly Supervised Action Detection
• STAR (AAAI19) Segregated Temporal Assembly Recurrent Networks for Weakly Supervised Multiple Action Detection
• 3C-Net (ICCV19) 3C-Net: Category Count and Center Loss for Weakly-Supervised Action Localization pytorch
• CMCS (CVPR19) Completeness Modeling and Context Separation for Weakly Supervised Temporal Action Localization pytorch
• MAAN (ICLR19) Marginalized Average Attentional Network for Weakly-Supervised Learning pytorch
• Nguyen et al. (ICCV19) Weakly-supervised Action Localization with Background Modeling
• CleanNet (ICCV19) Weakly Supervised Temporal Action Localization Through Contrast Based Evaluation Networks
• BaSNet (AAAI20) Background Suppression Network for Weakly-supervised Temporal Action Localization pytorch
• RPN (AAAI20) Relational Prototypical Network for Weakly Supervised Temporal Action Localization
• WSGN (WACV20) Weakly Supervised Gaussian Networks for Action Detection
• WSAD (WACV20) Weakly Supervised Temporal Action Localization Using Deep Metric Learning pytorch
• DGAM (CVPR20) Weakly-Supervised Action Localization by Generative Attention Modeling pytorch
• EM-MIL Weakly-Supervised Action Localization with Expectation-Maximization Multi-Instance Learning
• ACL Learning Temporal Co-Attention Models for Unsupervised Video Action Localization
• BMUE Background Modeling via Uncertainty Estimation for Weakly-supervised Action Localization pytorch

fully-supervised

• S-CNN (CVPR16) Temporal Action Localization in Untrimmed Videos via Multi-stage CNNs caffe
• P-GCN (ICCV19) Graph Convolutional Networks for Temporal Action Localization pytorch
• C-TCN (ICCV19) Deep Concept-wise Temporal Convolutional Networks for Action Localization
• Joshua et al. (WACV20) Activity Detection in Untrimmed Videos Using Chunk-based Classifiers

Experiments (THUMOS14)

• Thumos14 data download : link
• C3D feature : link
• I3D feature : link1(from here) or link2(from here)
  • extraction code : link2

# link2
├── THUMOS14
   ├── gt.json
   ├── split_train.txt
   ├── split_test.txt
   └── features
       ├── train
           ├── rgb
               ├── video_validation_0000051.npy
               ├── video_validation_0000052.npy
               └── ...
           └── flow
               ├── video_validation_0000051.npy
               ├── video_validation_0000052.npy
               └── ...
       └── test
           ├── rgb
               ├── video_test_0000004.npy
               ├── video_test_0000006.npy
               └── ...
           └── flow
               ├── video_test_0000004.npy
               ├── video_test_0000006.npy
               └── ...

• UNet : link1 or link2

weakly-supervised (feature extractor:I3D)

	conference	name	0.1	0.2	0.3	0.4	0.5	0.6	0.7	0.8	0.9	code
1	ECCV(2018)	W-TALC	55.2	49.6	40.1	31.1	22.8	-	7.6	-	-	tensor, pytorch
2	CVPR(2018)	STPN	52.0	44.7	35.5	25.8	16.9	9.9	4.3	1.2	0.1	tensor, pytorch
3	ACM(2019)	ASSG	65.6	59.4	50.4	38.7	25.4	15.0	6.6	-	-
4	ICCV(2019)	TSM	-	-	39.5	-	24.5	-	7.1	-
5	AAAI(2019)	STAR	68.8	60.0	48.7	34.7	23.0	-	-	-	-
6	ICCV(2019)	3C-Net	59.1	53.5	44.2	34.1	26.6	-	8.1	-	-	pytorch
7	CVPR(2019)	CMCS	57.4	50.8	41.2	32.1	23.1	15.0	7.0	-	-	pytorch
8	ICLR(2019)	MAAN	59.8	50.8	41.1	30.6	20.3	12.0	6.9	-	-	pytorch
9	ICCV(2019)	Nguyen et al.	60.4	56.0	46.6	37.5	26.8	17.6	9.0	3.3	0.4
10	ICCV(2019)	CleanNet	-	-	37.0	30.9	23.9	13.9	7.1	-	-
11	AAAI(2020)	BaSNet	58.2	52.3	44.6	36.0	27.0	18.6	10.4	3.9	0.5	pytorch
12	AAAI(2020)	RPN	62.3	57.0	48.2	37.2	27.9	16.7	8.1
13	WACV(2020)	WSGN
14	WACV(2020)	WSAD	62.3		46.8		29.6		9.7			pytorch
15	CVPR(2020)	DGAM	60.0	54.2	46.8	38.2	28.8	19.8	11.4	3.6	0.4	pytorch
16		EM-MIL	59.1	52.7	45.5	36.8	30.5	22.7	16.4	-	-
17		ACL	-	-	46.9	38.9	30.1	19.8	10.4	-
18		BMUE	-	-	46.9	39.2	30.7	20.8	12.5	-	-	pytorch

fully-supervised

	conference	name	0.1	0.2	0.3	0.4	0.5	0.6	0.7	0.8	0.9	code
1	CVPR(2016)	S-CNN	47.7	60.0	48.7	34.7	23.0	-	-	-	-	caffe
2	ICCV(2019)	P-GCN	69.5	67.8	63.6	57.8	49.1	-	-	-	-	pytorch
3	ICCV(2019)	C-TCN	72.2	71.4	68.0	62.3	52.1	-	-	-	-
4	WACV(2020)	Joshua et al.	67.41	60.0	48.7	34.7	23.0	-	-	-	-