PK PyTorchKR Papers
Papers with Code · CC BY-SA 4.0

Action Detection

566 평가 기록 62 datasets

Action Detection aims to find both where and when an action occurs within a video clip and classify what the action is taking place. Typically results are given in the form of action tublets, which are action bounding boxes linked across time in the video. This is related to temporal localization, which seeks to identify the start and end frame of an action, and action recognition, which seeks only to classify which action is taking place and typically assumes a trimmed video.

개요는 각 항목의 일부입니다. 위 탭에서 전체 기록을 페이지별로 탐색할 수 있습니다.

벤치마크

Papers with Code 아카이브의 데이터셋별 기록 예시입니다. 평가 조건과 지표 방향이 검증되지 않아 순위를 의미하지 않습니다.

Dataset기록된 모델 예시MetricPaperCode
Hulk(Finetune, ViT-L)
Accuracy (CS): 94.3
논문 코드
ProtoGCN
Accuracy (Cross-Subject): 90.9Accuracy (Cross-Setup): 92.2Ensembled Modalities: 6
논문 코드
Structured Keypoint Pooling (PPNv2 skeletons+objects)
Accuracy: 52.3
논문 -
DSCNet (RGB + Pose)
Accuracy: 99.1
논문 코드
LoCoNet+TalkNCE
validation mean average precision: 95.5%
논문 코드
STAR/L
Video-mAP 0.2: 88.0Video-mAP 0.5: 71.8Frame-mAP 0.5: 90.3
논문 -
SiA
Frame-mAP 0.5: 88.5
논문 코드
TTM
mAP: 28.79
논문 코드
GateHUB(TimeSformer)
mAP: 72.5
논문 -
Potion
Accuracy (RGB+pose): 90.4Accuracy (pose): 67.9
논문 -
GateHUB
mCAP: 89.6
논문 -
Joint Line Distance
Accuracy: 99.02%
논문 코드

데이터셋

62개
NTU RGB+DNTU RGB+D 120Kinetics-Skeleton datasetN-UCLAAVA-ActiveSpeakerUCF101-24J-HMDBCharadesTHUMOS'14J-HMDBTVSeriesSBU / SBU-Refine

논문

라이브러리

Papers with Code 리더보드 row에 연결된 구현 저장소입니다.

github.com/piergiaj/tgm-icml19 2 논문
github.com/boneddeng/tadml 1 논문
github.com/echo0125/memory-and-anticipation-transformer 1 논문
github.com/ccp-eva/sporttaskme22 1 논문
github.com/dairui01/MS-TCT 1 논문
github.com/dairui01/PDAN 1 논문
github.com/fidsinn/sporttaskme22 1 논문
github.com/piergiaj/super-events-cvpr18 1 논문

하위 태스크

SB Skeleton Based Action Recognition 34 데이터셋 · 415 rows OA Online Action Detection 3 데이터셋 · 29 rows AA Audio-Visual Active Speaker Detection 2 데이터셋 · 21 rows HA Human Activity Recognition 9 데이터셋 · 17 rows FS Few Shot Temporal Action Localization 2 데이터셋 · 2 rows AT Action Triplet Detection 1 데이터셋 · 1 rows FA Fine-Grained Action Detection 0 데이터셋 · 0 rows MA Multiple Action Detection 0 데이터셋 · 0 rows