Zero Shot Action Recognition On Hmdb51

Metriken

Top-1 Accuracy

Ergebnisse

Leistungsergebnisse verschiedener Modelle zu diesem Benchmark

		Paper Title
MOV (ViT-L/14)	64.7	Multimodal Open-Vocabulary Video Classification via Pre-Trained Vision and Language Models
OTI(ViT-L/14)	64	Orthogonal Temporal Interpolation for Zero-Shot Video Recognition
BIKE	61.4	Bidirectional Cross-Modal Knowledge Exploration for Video Recognition with Pre-trained Vision-Language Models
MOV (ViT-B/16)	60.8	Multimodal Open-Vocabulary Video Classification via Pre-Trained Vision and Language Models
IMP-MoE-L	59.1	Alternating Gradient Descent and Mixture-of-Experts for Integrated Multimodal Perception
VideoCoCa	58.7	VideoCoCa: Video-Text Modeling with Zero-Shot Transfer from Contrastive Captioners
Text4Vis	58.4	Revisiting Classifier: Transferring Vision-Language Models for Video Recognition
TC-CLIP	56.0	Leveraging Temporal Contextualization for Video Action Recognition
OST	55.9	OST: Refining Text Knowledge with Optimal Spatio-Temporal Descriptor for General Video Recognition
MAXI	52.3	MAtch, eXpand and Improve: Unsupervised Finetuning for Zero-Shot Action Recognition with Language Knowledge
VicTR (ViT-B/16)	51.0	VicTR: Video-conditioned Text Representations for Activity Recognition
LoCATe-GAT	50.7	LoCATe-GAT: Modeling Multi-Scale Local Context and Action Relationships for Zero-Shot Action Recognition
X-CLIP	44.6	Expanding Language-Image Pretrained Models for General Video Recognition
CLASTER	43.2	CLASTER: Clustering with Reinforcement Learning for Zero-Shot Action Recognition
ResT	41.1	Cross-modal Representation Learning for Zero-shot Action Recognition
AURL	39	Alignment-Uniformity aware Representation Learning for Zero-shot Video Classification
JigsawNet	38.7	Rethinking Zero-shot Action Recognition: Learning from Latent Atomic Actions
SPOT	35.9	Synthetic Sample Selection for Generalized Zero-Shot Learning
ER-ZSAR	35.3	Elaborative Rehearsal for Zero-shot Action Recognition
E2E	32.7	Rethinking Zero-shot Video Classification: End-to-end Training for Realistic Applications

0 of 29 row(s) selected.

Command Palette

Zero Shot Action Recognition On Hmdb51

Metriken

Ergebnisse