HyperAI초신경

홈 뉴스 최신 연구 논문 튜토리얼 데이터셋 백과사전 SOTA LLM 모델 GPU 랭킹 컨퍼런스

한국어

HyperAI초신경

Zero Shot Skeletal Action Recognition On Ntu

평가 지표

Accuracy (12 unseen classes)

Accuracy (5 unseen classes)

Random Split Accuracy

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	Accuracy (12 unseen classes)	Accuracy (5 unseen classes)	Random Split Accuracy	Paper Title	Repository
SA-DVAE	41.38	82.37	84.20	SA-DVAE: Improving Zero-Shot Skeleton-Based Action Recognition by Disentangled Variational Autoencoders	-
SA-DVAE + augmented text	-	-	87.61	SA-DVAE: Improving Zero-Shot Skeleton-Based Action Recognition by Disentangled Variational Autoencoders	-
PURLS	40.99	79.23	-	Part-aware Unified Representation of Language and Skeleton for Zero-shot Action Recognition	-
STAR	45.10	81.40	77.50	Fine-Grained Side Information Guided Dual-Prompts for Zero-Shot Skeleton Action Recognition	-
TDSM	56.03	86.49	88.88	TDSM: Triplet Diffusion for Skeleton-Text Matching in Zero-Shot Action Recognition	-
SMIE	40.18	77.98	65.08	Zero-shot Skeleton-based Action Recognition via Mutual Information Estimation and Maximization	-
CADA-VAE	28.96	76.84	60.74	Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders	-
SynSE	33.30	75.81	64.19	Syntactically Guided Generative Embeddings for Zero-Shot Skeleton Action Recognition	-

0 of 8 row(s) selected.

Zero Shot Skeletal Action Recognition On Ntu | SOTA | HyperAI초신경