SOTAVerified

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

Showing 126150 of 219 papers

TitleStatusHype
Fast and Unsupervised Action Boundary Detection for Action Segmentation0
You Can Wash Hands Better: Accurate Daily Handwashing Assessment with a SmartwatchCode0
Iterative Contrast-Classify For Semi-supervised Temporal Action SegmentationCode1
Towards Tokenized Human Dynamics RepresentationCode1
Few-Shot Temporal Action Localization with Query Adaptive TransformerCode1
ASFormer: Transformer for Action SegmentationCode1
Hierarchical Modeling for Task Recognition and Action Segmentation in Weakly-Labeled Instructional VideosCode0
VideoCLIP: Contrastive Pre-training for Zero-shot Video-Text Understanding0
Long Short View Feature Decomposition via Contrastive Video Representation Learning0
TACo: Token-aware Cascade Contrastive Learning for Video-Text Alignment0
Temporal Action Segmentation with High-level Complex Activity Labels0
FIFA: Fast Inference Approximation for Action Segmentation0
Unsupervised Action Segmentation for Instructional Videos0
SSCAP: Self-supervised Co-occurrence Action Parsing for Unsupervised Temporal Action Segmentation0
Unsupervised Action Segmentation by Joint Representation Learning and Online ClusteringCode1
Coarse to Fine Multi-Resolution Temporal Convolutional NetworkCode1
VLM: Task-agnostic Video-Language Model Pre-training for Video Understanding0
Efficient Two-Step Networks for Temporal Action SegmentationCode1
Unsupervised Discriminative Embedding for Sub-Action Learning in Complex Activities0
Action in Mind: A Neural Network Approach to Action Recognition and Segmentation0
Action Segmentation with Mixed Temporal Domain Adaptation0
Anchor-Constrained Viterbi for Set-Supervised Action Segmentation0
Action Shuffle Alternating Learning for Unsupervised Action Segmentation0
Automated freezing of gait assessment with marker-based motion capture and multi-stage spatial-temporal graph convolutional neural networksCode1
Temporally-Weighted Hierarchical Clustering for Unsupervised Action SegmentationCode1
Show:102550
← PrevPage 6 of 9Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1AdaFocus (newly extracted I3D-features, LT-Context model)Average F176.2Unverified
2FACT (efficient hybrid of convolution and transformer model)Average F174.7Unverified
3ASQueryAverage F174.6Unverified
4BITAverage F173.7Unverified
5DiffActAverage F173.6Unverified
6BaFormerAverage F172.4Unverified
7CETNetAverage F171.8Unverified
8SF-TMN(ASFormer)Average F171.6Unverified
9RF++-SSTDAAcc70.8Unverified
10ASPnetAverage F170.6Unverified
#ModelMetricClaimedVerifiedStatus
1Br-Prompt+ASPnet (RGB, flow, accelerometer)F1@50%88.5Unverified
2Semantic2GraphF1@50%87.3Unverified
3BaFormerF1@50%83.9Unverified
4DiffActF1@50%83.7Unverified
5SF-TMN(ASFormer)F1@50%82.9Unverified
6LTContextF1@50%82Unverified
7UVASTF1@50%81.7Unverified
8Br-Prompt+ASFormerF1@50%81.3Unverified
9EUTF1@50%81Unverified
10CETNetF1@50%80.1Unverified
#ModelMetricClaimedVerifiedStatus
1Semantic2GraphF1@50%91.3Unverified
2FACTF1@50%87.5Unverified
3DiffActF1@50%84.7Unverified
4BaFormerF1@50%83.5Unverified
5SF-TMN(ASFormer)F1@50%83.1Unverified
6Br-Prompt+ASFormerF1@50%83Unverified
7DPRNF1@50%82.9Unverified
8BITF1@50%82.6Unverified
9CETNetF1@50%81.3Unverified
10UVASTF1@50%81Unverified
#ModelMetricClaimedVerifiedStatus
1UnLoc-LFrame accuracy72.8Unverified
2UnivlFrame accuracy70Unverified
3NortonFrame accuracy69.8Unverified
4VideoClipFrame accuracy68.7Unverified
5TACoFrame accuracy68.4Unverified
6VLMFrame accuracy68.4Unverified
7MIL-NCEFrame accuracy61Unverified
8ActBERTFrame accuracy57Unverified
9CBTFrame accuracy53.9Unverified
#ModelMetricClaimedVerifiedStatus
1ASQueryF1@10%37.8Unverified
2LTContextF1@10%33.9Unverified
3ASFormerF1@10%33.4Unverified
4C2F-TCNF1@10%33.3Unverified
5UVASTF1@10%32.1Unverified
6MS-TCN++F1@10%31.6Unverified
7ProTAS(Offline)F1@10%28.7Unverified
#ModelMetricClaimedVerifiedStatus
1RL+TreeEdit Distance88.53Unverified
2RL (full)Edit Distance87.96Unverified
3TricorNetEdit Distance86.8Unverified
4SDL+SC-CRFEdit Distance86.21Unverified
5TCNEdit Distance83.1Unverified
6ST-CNN+SegEdit Distance66.56Unverified
#ModelMetricClaimedVerifiedStatus
1TSA (FINCH)Acc62.4Unverified
2TSA (Kmeans)Acc59.7Unverified
#ModelMetricClaimedVerifiedStatus
1EUTAcc87.4Unverified
#ModelMetricClaimedVerifiedStatus
1Unsup. TW-FINCH (K=avg/activity)Accuracy42Unverified