Action Segmentation

Action Segmentation is a challenging problem in high-level video understanding. In its simplest form, Action Segmentation aims to segment a temporally untrimmed video by time and label each segmented part with one of pre-defined action labels. The results of Action Segmentation can be further used as input to various applications, such as video-to-text and action localization.

Source: TricorNet: A Hybrid Temporal Convolutional and Recurrent Network for Video Action Segmentation

Papers

Recently Added Most Hyped Most Active Needs Verification Most Verified

Showing 201–219 of 219 papers

Title	Date	Tasks	Status	Hype
Actor and Action Video Segmentation from a Sentence	Mar 20, 2018	Action SegmentationDecoder	CodeCode Available	1
Temporal Human Action Segmentation via Dynamic Clustering	Mar 15, 2018	Action SegmentationClustering	CodeCode Available	0
End-to-End Fine-Grained Action Segmentation and Recognition Using Conditional Random Field Models and Discriminative Sparse Coding	Jan 29, 2018	Action ClassificationAction Segmentation	—Unverified	0
Video Action Segmentation with Hybrid Temporal Networks	Jan 1, 2018	Action SegmentationDecoder	—Unverified	0
An End-to-end 3D Convolutional Neural Network for Action Detection and Segmentation in Videos	Nov 30, 2017	Action DetectionAction Segmentation	—Unverified	0
Weakly Supervised Actor-Action Segmentation via Robust Multi-Task Ranking	Jul 1, 2017	Action ClassificationAction Segmentation	—Unverified	0
Action Sets: Weakly Supervised Action Segmentation without Ordering Constraints	Jun 2, 2017	Action DetectionAction Segmentation	CodeCode Available	0
TricorNet: A Hybrid Temporal Convolutional and Recurrent Network for Video Action Segmentation	May 22, 2017	Action SegmentationDecoder	—Unverified	0
Action Understanding with Multiple Classes of Actors	Apr 27, 2017	Action RecognitionAction Segmentation	—Unverified	0
Weakly Supervised Action Learning with RNN based Fine-to-coarse Modeling	Mar 23, 2017	Action SegmentationTemporal Action Segmentation	CodeCode Available	0
Temporal Convolutional Networks for Action Segmentation and Detection	Nov 16, 2016	Action SegmentationDecoder	CodeCode Available	1
Temporal Convolutional Networks: A Unified Approach to Action Segmentation	Aug 29, 2016	Action SegmentationSegmentation	CodeCode Available	0
Watch-n-Patch: Unsupervised Learning of Actions and Relations	Mar 11, 2016	Action SegmentationClustering	—Unverified	0
Segmental Spatiotemporal CNNs for Fine-grained Action Segmentation	Feb 9, 2016	Action ClassificationAction Recognition	—Unverified	0
Watch-Bot: Unsupervised Learning for Reminding Humans of Forgotten Actions	Dec 14, 2015	Action SegmentationObject	—Unverified	0
Grasp Type Revisited: A Modern Perspective on a Classical Feature for Vision	Jun 1, 2015	Action SegmentationAction Understanding	—Unverified	0
Watch-n-Patch: Unsupervised Understanding of Actions and Relations	Jun 1, 2015	Action SegmentationUnsupervised Action Segmentation	—Unverified	0
Human Action Segmentation With Hierarchical Supervoxel Consistency	Jun 1, 2015	Action ClassificationAction Segmentation	—Unverified	0
Leveraging Hierarchical Parametric Networks for Skeletal Joints Based Action Segmentation and Recognition	Jun 1, 2014	Action RecognitionAction Segmentation	—Unverified	0

Show:10 25 50

← PrevPage 5 of 5Next →

All datasets Breakfast 50 Salads GTEA COIN Assembly101 JIGSAWS Youtube INRIA Instructional 50Salads MPII Cooking 2 Dataset

Benchmark Results

#	Model	Metric	Claimed	Verified	Status
1	AdaFocus (newly extracted I3D-features, LT-Context model)	Average F1	76.2	—	Unverified
2	FACT (efficient hybrid of convolution and transformer model)	Average F1	74.7	—	Unverified
3	ASQuery	Average F1	74.6	—	Unverified
4	BIT	Average F1	73.7	—	Unverified
5	DiffAct	Average F1	73.6	—	Unverified
6	BaFormer	Average F1	72.4	—	Unverified
7	CETNet	Average F1	71.8	—	Unverified
8	SF-TMN(ASFormer)	Average F1	71.6	—	Unverified
9	RF++-SSTDA	Acc	70.8	—	Unverified
10	ASPnet	Average F1	70.6	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Br-Prompt+ASPnet (RGB, flow, accelerometer)	F1@50%	88.5	—	Unverified
2	Semantic2Graph	F1@50%	87.3	—	Unverified
3	BaFormer	F1@50%	83.9	—	Unverified
4	DiffAct	F1@50%	83.7	—	Unverified
5	SF-TMN(ASFormer)	F1@50%	82.9	—	Unverified
6	LTContext	F1@50%	82	—	Unverified
7	UVAST	F1@50%	81.7	—	Unverified
8	Br-Prompt+ASFormer	F1@50%	81.3	—	Unverified
9	EUT	F1@50%	81	—	Unverified
10	CETNet	F1@50%	80.1	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Semantic2Graph	F1@50%	91.3	—	Unverified
2	FACT	F1@50%	87.5	—	Unverified
3	DiffAct	F1@50%	84.7	—	Unverified
4	BaFormer	F1@50%	83.5	—	Unverified
5	SF-TMN(ASFormer)	F1@50%	83.1	—	Unverified
6	Br-Prompt+ASFormer	F1@50%	83	—	Unverified
7	DPRN	F1@50%	82.9	—	Unverified
8	BIT	F1@50%	82.6	—	Unverified
9	CETNet	F1@50%	81.3	—	Unverified
10	UVAST	F1@50%	81	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	UnLoc-L	Frame accuracy	72.8	—	Unverified
2	Univl	Frame accuracy	70	—	Unverified
3	Norton	Frame accuracy	69.8	—	Unverified
4	VideoClip	Frame accuracy	68.7	—	Unverified
5	TACo	Frame accuracy	68.4	—	Unverified
6	VLM	Frame accuracy	68.4	—	Unverified
7	MIL-NCE	Frame accuracy	61	—	Unverified
8	ActBERT	Frame accuracy	57	—	Unverified
9	CBT	Frame accuracy	53.9	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	ASQuery	F1@10%	37.8	—	Unverified
2	LTContext	F1@10%	33.9	—	Unverified
3	ASFormer	F1@10%	33.4	—	Unverified
4	C2F-TCN	F1@10%	33.3	—	Unverified
5	UVAST	F1@10%	32.1	—	Unverified
6	MS-TCN++	F1@10%	31.6	—	Unverified
7	ProTAS(Offline)	F1@10%	28.7	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	RL+Tree	Edit Distance	88.53	—	Unverified
2	RL (full)	Edit Distance	87.96	—	Unverified
3	TricorNet	Edit Distance	86.8	—	Unverified
4	SDL+SC-CRF	Edit Distance	86.21	—	Unverified
5	TCN	Edit Distance	83.1	—	Unverified
6	ST-CNN+Seg	Edit Distance	66.56	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	TSA (FINCH)	Acc	62.4	—	Unverified
2	TSA (Kmeans)	Acc	59.7	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	EUT	Acc	87.4	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Unsup. TW-FINCH (K=avg/activity)	Accuracy	42	—	Unverified