SOTAVerified

Action Recognition

Action Recognition is a computer vision task that involves recognizing human actions in videos or images. The goal is to classify and categorize the actions being performed in the video or image into a predefined set of action classes.

In the video domain, it is an open question whether training an action classification network on a sufficiently large dataset, will give a similar boost in performance when applied to a different temporal task or dataset. The challenges of building video datasets has meant that most popular benchmarks for action recognition are small, having on the order of 10k videos.

Please note some benchmarks may be located in the Action Classification or Video Classification tasks, e.g. Kinetics-400.

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 151–200 of 2759 papers

TitleStatusHype
CDFSL-V: Cross-Domain Few-Shot Learning for VideosCode1
CHASE: Learning Convex Hull Adaptive Shift for Skeleton-based Multi-Entity Action RecognitionCode1
Actor-Context-Actor Relation Network for Spatio-Temporal Action LocalizationCode1
C2C: Component-to-Composition Learning for Zero-Shot Compositional Action RecognitionCode1
CAKES: Channel-wise Automatic KErnel Shrinking for Efficient 3D NetworksCode1
CAST: Cross-Attention in Space and Time for Video Action RecognitionCode1
Full-Body Articulated Human-Object InteractionCode1
Challenges in Video-Based Infant Action Recognition: A Critical Examination of the State of the ArtCode1
CMD: Self-supervised 3D Action Representation Learning with Cross-modal Mutual DistillationCode1
Building a Multi-modal Spatiotemporal Expert for Zero-shot Action Recognition with CLIPCode1
CoFInAl: Enhancing Action Quality Assessment with Coarse-to-Fine Instruction AlignmentCode1
Collaborating Domain-shared and Target-specific Feature Clustering for Cross-domain 3D Action RecognitionCode1
A Large-scale Study of Spatiotemporal Representation Learning with a New Benchmark on Action RecognitionCode1
3DV: 3D Dynamic Voxel for Action Recognition in Depth VideoCode1
Action Genome: Actions as Composition of Spatio-temporal Scene GraphsCode1
Concatenated Masked Autoencoders as Spatial-Temporal LearnerCode1
3DYoga90: A Hierarchical Video Dataset for Yoga Pose UnderstandingCode1
Context-Aware RCNN: A Baseline for Action Detection in VideosCode1
Contrastive Learning from Spatio-Temporal Mixed Skeleton Sequences for Self-Supervised Skeleton-Based Action RecognitionCode1
ConvNet Architecture Search for Spatiotemporal Feature LearningCode1
Cross-Architecture Self-supervised Video Representation LearningCode1
A Deeper Dive Into What Deep Spatiotemporal Networks Encode: Quantifying Static vs. Dynamic InformationCode1
CZU-MHAD: A multimodal dataset for human action recognition utilizing a depth camera and 10 wearable inertial sensorsCode1
A Dense-Sparse Complementary Network for Human Action Recognition based on RGB and Skeleton ModalitiesCode1
Action knowledge for video captioning with graph neural networksCode1
Decoupled Spatial-Temporal Attention Network for Skeleton-Based Action RecognitionCode1
Deep Analysis of CNN-based Spatio-temporal Representations for Action RecognitionCode1
Building an Open-Vocabulary Video CLIP Model with Better Architectures, Optimization and DataCode1
CIAGAN: Conditional Identity Anonymization Generative Adversarial NetworksCode1
BMN: Boundary-Matching Network for Temporal Action Proposal GenerationCode1
Depth Guided Adaptive Meta-Fusion Network for Few-shot Video RecognitionCode1
DEVIAS: Learning Disentangled Video Representations of Action and SceneCode1
ActionCLIP: A New Paradigm for Video Action RecognitionCode1
Disentangled Pre-training for Human-Object Interaction DetectionCode1
Diverse Temporal Aggregation and Depthwise Spatiotemporal Factorization for Efficient Video ClassificationCode1
Do Language Models Understand Time?Code1
ACTION-Net: Multipath Excitation for Action RecognitionCode1
DSANet: Dynamic Segment Aggregation Network for Video-Level Representation LearningCode1
Bridging Video-text Retrieval with Multiple Choice QuestionsCode1
Dual-path Adaptation from Image to Video TransformersCode1
Blindly Assess Quality of In-the-Wild Videos via Quality-aware Pre-training and Motion PerceptionCode1
An Action Is Worth Multiple Words: Handling Ambiguity in Action RecognitionCode1
Bringing Online Egocentric Action Recognition into the wildCode1
EgoAdapt: A multi-stream evaluation study of adaptation to real-world egocentric user videoCode1
Benchmarking Micro-action Recognition: Dataset, Methods, and ApplicationsCode1
EgoSurgery-Phase: A Dataset of Surgical Phase Recognition from Egocentric Open Surgery VideosCode1
EgoVLPv2: Egocentric Video-Language Pre-training with Fusion in the BackboneCode1
Elaborative Rehearsal for Zero-shot Action RecognitionCode1
Enhancing Unsupervised Video Representation Learning by Decoupling the Scene and the MotionCode1
BASAR:Black-box Attack on Skeletal Action RecognitionCode1
Show:102550
← PrevPage 4 of 56Next →
All datasetsSomething-Something V2UCF101HMDB-51Something-Something V1AVA v2.2EPIC-KITCHENS-100NTU RGB+DNTU RGB+D 120Diving-48ActivityNetAVA v2.1H2O (2 Hands and Objects)

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1MViTv2-B (IN-21K + Kinetics400 pretrain)Top-5 Accuracy93.4—Unverified
2RSANet-R50 (8+16 frames, ImageNet pretrained, 2 clips)Top-5 Accuracy91.1—Unverified
3MVD (Kinetics400 pretrain, ViT-H, 16 frame)Top-1 Accuracy77.3—Unverified
4InternVideoTop-1 Accuracy77.2—Unverified
5DejaVidTop-1 Accuracy77.2—Unverified
6InternVideo2-1BTop-1 Accuracy77.1—Unverified
7VideoMAE V2-gTop-1 Accuracy77—Unverified
8MVD (Kinetics400 pretrain, ViT-L, 16 frame)Top-1 Accuracy76.7—Unverified
9Hiera-L (no extra data)Top-1 Accuracy76.5—Unverified
10TubeViT-LTop-1 Accuracy76.1—Unverified
#ModelMetricClaimedVerifiedStatus
1FTP-UniFormerV2-L/143-fold Accuracy99.7—Unverified
2OmniVec3-fold Accuracy99.6—Unverified
3VideoMAE V2-g3-fold Accuracy99.6—Unverified
4OmniVec23-fold Accuracy99.6—Unverified
5BIKE3-fold Accuracy98.8—Unverified
6SMART3-fold Accuracy98.64—Unverified
7ZeroI2V ViT-L/143-fold Accuracy98.6—Unverified
8PERF-Net (multi-distilled S3D)3-fold Accuracy98.6—Unverified
9OmniSource (SlowOnly-8x8-R101-RGB + I3D-Flow)3-fold Accuracy98.6—Unverified
10Text4Vis3-fold Accuracy98.2—Unverified