Action Classification On Kinetics 400

评估指标

Acc@1

评测结果

各个模型在此基准测试上的表现结果

		Paper Title	Repository
OmniVec2	93.6	OmniVec2 - A Novel Transformer based Network for Large Scale Multimodal and Multitask Learning	-
FTP-UniFormerV2-L/14	93.4	Enhancing Video Transformers for Action Understanding with VLM-aided Training	-
InternVideo2-6B	92.1	InternVideo2: Scaling Foundation Models for Multimodal Video Understanding
InternVideo2-1B	91.6	InternVideo2: Scaling Foundation Models for Multimodal Video Understanding
OmniVec	91.1	OmniVec: Learning robust representations with cross modal sharing	-
InternVideo	91.1	InternVideo: General Video Foundation Models via Generative and Discriminative Learning
TubeViT-H (ImageNet-1k)	90.9	Rethinking Video ViTs: Sparse Video Tubes for Joint Image and Video Learning
UMT-L (ViT-L/16)	90.6	Unmasked Teacher: Towards Training-Efficient Video Foundation Models
Unmasked Teacher (ViT-L)	90.6	Unmasked Teacher: Towards Training-Efficient Video Foundation Models
TubeVit-L (ImageNet-1k)	90.2	Rethinking Video ViTs: Sparse Video Tubes for Joint Image and Video Learning
UniFormerV2-L (ViT-L, 336)	90.0	UniFormerV2: Spatiotemporal Learning by Arming Image ViTs with Video UniFormer	-
VideoMAE V2-g (64x266x266)	90.0	VideoMAE V2: Scaling Video Masked Autoencoders with Dual Masking
MTV-H (WTS 60M)	89.9	Multiview Transformers for Video Recognition
TAdaFormer-L/14	89.9	Temporally-Adaptive Models for Efficient Video Understanding
EVA	89.7	EVA: Exploring the Limits of Masked Visual Representation Learning at Scale
AM/12 ViT-B Dinov2	89.6	AM Flow: Adapters for Temporal Processing in Action Recognition	-
ATM	89.4	What Can Simple Arithmetic Operations Do for Temporal Modeling?
CoCa (finetuned)	88.9	CoCa: Contrastive Captioners are Image-Text Foundation Models
ILA (ViT-L/14)	88.7	Implicit Temporal Modeling with Learnable Alignment for Video Recognition
BIKE (CLIP ViT-L/14)	88.7	Bidirectional Cross-Modal Knowledge Exploration for Video Recognition with Pre-trained Vision-Language Models

0 of 204 row(s) selected.