Papers with Code · CC BY-SA 4.0

Something-Something V2

아카이브 평가 123건 · 마지막 적재/변경 2026-10-30

평가 기록

과거 수집 기록이며 최신 순위가 아닙니다. 데이터셋 버전·split·실험 조건과 지표 방향이 검증되지 않은 결과는 직접 비교할 수 없습니다. 날짜는 논문 날짜이며 최신 평가일을 뜻하지 않습니다.

표가 화면보다 넓으면 좌우로 스크롤하여 논문과 구현 링크를 확인하세요.

원본 기록 순서 · 순위 미부여
모델	원본 지표·값	논문	구현
SELFYNet-TSM-R50En (8+16 frames, ImageNet pretrained, 2 clips)	Top-1 Accuracy 67.7 Top-5 Accuracy 91.1	Learning Self-Similarity in Space and Time as Generalized Motion for Video Action Recognition 2021-02-14	코드 1
RSANet-R50 (8+16 frames, ImageNet pretrained, 2 clips	Top-1 Accuracy 67.7 Top-5 Accuracy 91.1	Relational Self-Attention: What's Missing in Attention for Video Understanding 2021-11-02	코드 1
GTDNet	Top-1 Accuracy 67.6	Global Temporal Difference Network for Action Recognition 2022-11-23	연결 없음
SELFYNet-TSM-R50En (8+16 frames, ImageNet pretrained, a single clip)	Top-1 Accuracy 67.4 Top-5 Accuracy 91	Learning Self-Similarity in Space and Time as Generalized Motion for Video Action Recognition 2021-02-14	코드 1
VoV3D-L (32frames, Kinetics pretrained, single)	Top-1 Accuracy 67.35 Top-5 Accuracy 90.50 Parameters 5.8M GFLOPs 20.9x6	Diverse Temporal Aggregation and Depthwise Spatiotemporal Factorization for Efficient Video Classification 2020-12-01	코드 1
PLAR	Top-1 Accuracy 67.3 Top-5 Accuracy 91	SCP: Soft Conditional Prompt Learning for Aerial Video Action Recognition 2023-05-21	연결 없음
RSANet-R50 (8+16 frames, ImageNet pretrained, a single clip)	Top-1 Accuracy 67.3 Top-5 Accuracy 90.8	Relational Self-Attention: What's Missing in Attention for Video Understanding 2021-11-02	코드 1
X-Vit (x16)	Top-1 Accuracy 67.2 Top-5 Accuracy 90.8 Parameters N/A GFLOPs 850x1	Space-time Mixing Attention for Video Transformer 2021-06-10	코드 1
TAda2D-En (ResNet-50, 8+16 frames)	Top-1 Accuracy 67.2 Top-5 Accuracy 89.8	TAda! Temporally-Adaptive Convolutions for Video Understanding 2021-10-12	코드 1 · 코드 2
Mformer-HR	Top-1 Accuracy 67.1 Top-5 Accuracy 90.6 Parameters N/A GFLOPs 958.8x3	Keeping Your Eye on the Ball: Trajectory Attention in Video Transformers 2021-06-09	코드 1 · 코드 2
TAdaConvNeXt-T	Top-1 Accuracy 67.1 Top-5 Accuracy 90.4	TAda! Temporally-Adaptive Convolutions for Video Understanding 2021-10-12	코드 1 · 코드 2
MoDS (8+16frames)	Top-1 Accuracy 67.1	Action Recognition With Motion Diversification and Dynamic Selection 2022-07-15	연결 없음
STPG (8+16frames)	Top-1 Accuracy 67.0	Spatial-Temporal Pyramid Graph Reasoning for Action Recognition 2022-08-09	연결 없음
MML (single)	Top-1 Accuracy 66.83 Top-5 Accuracy 91.30	Mutual Modality Learning for Video Action Classification 2020-11-04	코드 1
ILA (ViT-B/16)	Top-1 Accuracy 66.8 Top-5 Accuracy 90.3	Implicit Temporal Modeling with Learnable Alignment for Video Recognition 2023-04-20	코드 1
TSM (RGB + Flow)	Top-1 Accuracy 66.6 Top-5 Accuracy 91.3	TSM: Temporal Shift Module for Efficient Video Understanding 2018-11-20	코드 1 · 코드 2 · 코드 3 · 코드 4 · 코드 5 · 코드 6 · 코드 7 · 코드 8 · 코드 9 · 코드 10 · 코드 11 · 코드 12 · 코드 13
MSNet-R50En (8+16 ensemble, ImageNet pretrained)	Top-1 Accuracy 66.6 Top-5 Accuracy 90.6	MotionSqueeze: Neural Motion Feature Learning for Video Understanding 2020-07-20	코드 1 · 코드 2
PAN ResNet101 (RGB only, no Flow)	Top-1 Accuracy 66.5 Top-5 Accuracy 90.6	PAN: Towards Fast Action Recognition via Learning Persistence of Appearance 2020-08-08	코드 1 · 코드 2
TSM+W3 (16 frames, RGB ResNet-50)	Top-1 Accuracy 66.5 Top-5 Accuracy 90.4	Knowing What, Where and When to Look: Efficient Video Action Modeling with Attention 2020-04-02	연결 없음
Mformer	Top-1 Accuracy 66.5 Top-5 Accuracy 90.1	Keeping Your Eye on the Ball: Trajectory Attention in Video Transformers 2021-06-09	코드 1 · 코드 2
MVFNet-ResNet50 (center crop, 8+16 ensemble, ImageNet pretrained, RGB only)	Top-1 Accuracy 66.3	MVFNet: Multi-View Fusion Network for Efficient Video Recognition 2020-12-13	코드 1 · 코드 2 · 코드 3
MViT-B, 16x4	Top-1 Accuracy 66.2 Top-5 Accuracy 90.2	Multiscale Vision Transformers 2021-04-22	코드 1 · 코드 2 · 코드 3 · 코드 4 · 코드 5 · 코드 6 · 코드 7 · 코드 8
RSANet-R50 (16 frames, ImageNet pretrained, a single clip)	Top-1 Accuracy 66 Top-5 Accuracy 89.8	Relational Self-Attention: What's Missing in Attention for Video Understanding 2021-11-02	코드 1
VoV3D-L (32frames, from scratch, single)	Top-1 Accuracy 65.8 Top-5 Accuracy 89.5 Parameters 5.8M GFLOPs 20.9x6	Diverse Temporal Aggregation and Depthwise Spatiotemporal Factorization for Efficient Video Classification 2020-12-01	코드 1
E3D-L	Top-1 Accuracy 65.7 Top-5 Accuracy 89.8 GFLOPs 18.3	Maximizing Spatio-Temporal Entropy of Deep 3D CNNs for Efficient Video Recognition 2023-03-05	코드 1
SELFYNet-TSM-R50 (16 frames, ImageNet pretrained)	Top-1 Accuracy 65.7 Top-5 Accuracy 89.8	Learning Self-Similarity in Space and Time as Generalized Motion for Video Action Recognition 2021-02-14	코드 1
TAda2D (ResNet-50, 16 frames)	Top-1 Accuracy 65.6 Top-5 Accuracy 89.2	TAda! Temporally-Adaptive Convolutions for Video Understanding 2021-10-12	코드 1 · 코드 2
ViViT-L/16x2 Fact. encoder	Top-1 Accuracy 65.4 Top-5 Accuracy 89.8	ViViT: A Video Vision Transformer 2021-03-29	코드 1 · 코드 2 · 코드 3 · 코드 4 · 코드 5 · 코드 6 · 코드 7 · 코드 8 · 코드 9 · 코드 10
VoV3D-M (32frames, Kinetics pretrained, single)	Top-1 Accuracy 65.24 Top-5 Accuracy 89.48 Parameters 3.3M GFLOPs 11.5x6	Diverse Temporal Aggregation and Depthwise Spatiotemporal Factorization for Efficient Video Classification 2020-12-01	코드 1
bLVNet	Top-1 Accuracy 65.2	More Is Less: Learning Efficient Video Representations by Big-Little Network and Depthwise Temporal Aggregation 2019-12-02	코드 1