Zero Shot Action Recognition On Ucf101

評価指標

Top-1 Accuracy

評価結果

このベンチマークにおける各モデルのパフォーマンス結果

モデル名	Top-1 Accuracy	Paper Title	Repository
SVE	10.9	Semantic Embedding Space for Zero-Shot Action Recognition	-
ResT	58.7	Cross-modal Representation Learning for Zero-shot Action Recognition	-
SJE(Attribute)	12.0	Evaluation of Output Embeddings for Fine-Grained Image Classification
MAXI	78.2	MAtch, eXpand and Improve: Unsupervised Finetuning for Zero-Shot Action Recognition with Language Knowledge
TC-CLIP	85.4	Leveraging Temporal Contextualization for Video Action Recognition
BIKE	86.6	Bidirectional Cross-Modal Knowledge Exploration for Video Recognition with Pre-trained Vision-Language Models
MOV (ViT-B/16)	82.6	Multimodal Open-Vocabulary Video Classification via Pre-Trained Vision and Language Models	-
O2A	30.3	Objects2action: Classifying and localizing actions without any video example	-
X-CLIP	72.0	Expanding Language-Image Pretrained Models for General Video Recognition
ZSECOC	15.1	Zero-Shot Action Recognition With Error-Correcting Output Codes	-
Text4Vis	85.8	Revisiting Classifier: Transferring Vision-Language Models for Video Recognition
VicTR (ViT-B/16)	72.4	VicTR: Video-conditioned Text Representations for Activity Recognition	-
LoCATe-GAT	76.0	LoCATe-GAT: Modeling Multi-Scale Local Context and Action Relationships for Zero-Shot Action Recognition
ESZSL	15.0	An embarrassingly simple approach to zero-shot learning
HAA	14.9	-	-
VideoCoCa	86.6	VideoCoCa: Video-Text Modeling with Zero-Shot Transfer from Contrastive Captioners	-
AURL	58	Alignment-Uniformity aware Representation Learning for Zero-shot Video Classification	-
CLASTER	53.9	CLASTER: Clustering with Reinforcement Learning for Zero-Shot Action Recognition	-
EVA-CLIP-E/14+	83.1	EVA-CLIP: Improved Training Techniques for CLIP at Scale
TS-GCN	34.2	I Know the Relationships: Zero-Shot Action Recognition via Two-Stream Graph Convolutional Networks and Knowledge Graphs

0 of 35 row(s) selected.