Papers with Code 아카이브

Action Classification

태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.

전체 기법 전체 데이터셋

하위 태스크: 0
평가표: 28
보존된 평가 기록: 444

Image source: The Kinetics Human Action Video Dataset

코드

관련 논문과 평가 결과에 연결된 구현입니다. 출처에서 확인된 저자 구현에는 공식 코드 표시가 있습니다.

https://github.com/yiyixuxu/TimeSformer-rolled-attention

Is Space-Time Attention All You Need for Video Understanding?

https://github.com/youngjun0627/movie-rating

SlowFast Networks for Video Recognition

https://github.com/ytw1996/swin3d_mindspore

Video Swin Transformer

https://github.com/yyk-wew/semanticmim

Masked Feature Prediction for Self-Supervised Visual Pre-Training

https://github.com/zhengshou/AutoLoc

UntrimmedNets for Weakly Supervised Action Recognition and Detection

https://github.com/zhenyangli/VideoLSTM

VideoLSTM Convolves, Attends and Flows for Action Recognition

https://github.com/zhoubolei/TRN-pytorch

Temporal Relational Reasoning in Videos

https://github.com/zrl4836/tensorflow_octConv

Drop an Octave: Reducing Spatial Redundancy in Convolutional Neural Networks with Octave Convolution

https://gitlab.com/birder/birder

Swin Transformer V2: Scaling Up Capacity and Resolution

출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.