Papers with Code 아카이브
Action Classification
태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.
- 하위 태스크
- 0
- 평가표
- 28
- 보존된 평가 기록
- 444
Image source: The Kinetics Human Action Video Dataset
논문
이 태스크와 하위 분야의 관련 논문입니다. 실제 성능 비교는 평가 결과에서 확인하세요.
An Empirical Study of End-to-End Temporal Action Detection
Deformable Video Transformer
SPAct: Self-supervised Privacy Preservation for Action Recognition
Frame-wise Action Representations for Long Videos via Sequence Contrastive Learning
VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training
Point3D: tracking actions as moving points with 3D CNNs
DirecFormer: A Directed Attention in Transformer Approach to Robust Action Recognition
Know your sensORs -- A Modality Study For Surgical Action Classification
OpenTAL: Towards Open Set Temporal Action Localization
Quantification of Occlusion Handling Capability of a 3D Human Pose Estimation Framework
Delving Deep into One-Shot Skeleton-based Action Recognition with Diverse Occlusions
Vision Models Are More Robust And Fair When Pretrained On Uncurated Images Without Supervision
Learning To Recognize Procedural Activities with Distant Supervision
End-to-end Generative Pretraining for Multimodal Video Captioning
Omnivore: A Single Model for Many Visual Modalities
MeMViT: Memory-Augmented Multiscale Vision Transformer for Efficient Long-Term Video Recognition
Video Transformers: A Survey
Multiview Transformers for Video Recognition
MERLOT Reserve: Neural Script Knowledge through Vision and Language and Sound
Masked Feature Prediction for Self-Supervised Visual Pre-Training
Spatio-Temporal CNN baseline method for the Sports Video Task of MediaEval 2021 benchmark
Co-training Transformer with Videos and Images Improves Action Recognition
Self-supervised Video Transformer
MViTv2: Improved Multiscale Vision Transformers for Classification and Detection
PreViTS: Contrastive Pretraining with Video Tracking Supervision
Hierarchical Graph-Convolutional Variational AutoEncoding for Generative Modelling of Human Motion
Florence: A New Foundation Model for Computer Vision
Swin Transformer V2: Scaling Up Capacity and Resolution
Co-segmentation Inspired Attention Module for Video-based Computer Vision Tasks
Revisiting spatio-temporal layouts for compositional action recognition
NoisyActions2M: A Multimedia Dataset for Video Understanding from Noisy Labels
TAda! Temporally-Adaptive Convolutions for Video Understanding
Three-Stream 3D/1D CNN for Fine-Grained Action Classification and Segmentation in Table Tennis
Class incremental learning for video action classification
ActionCLIP: A New Paradigm for Video Action Recognition
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.