HyperAI초신경

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

한국어

HyperAI초신경

Image Classification On Places365

평가 지표

Top 1 Accuracy

평가 결과

이 벤치마크에서 각 모델의 성능 결과

		Paper Title	Repository
OmniVec2	65.1	OmniVec2 - A Novel Transformer based Network for Large Scale Multimodal and Multitask Learning	-
OmniVec(ViT)	63.5	OmniVec: Learning robust representations with cross modal sharing	-
InternImage-H（CNN）	61.2%	InternImage: Exploring Large-Scale Vision Foundation Models with Deformable Convolutions
MixMIM-L(ViT-L)	60.3	MixMAE: Mixed and Masked Autoencoder for Efficient Pretraining of Hierarchical Vision Transformers
ViC-MAE (ViT-L)	59.5%	ViC-MAE: Self-Supervised Representation Learning from Images and Video with Contrastive Masked Autoencoders
µ2Net+ (ViT-L/16)	59.15	A Continual Development Methodology for Large-scale Multitask Dynamic ML Systems
MixMIM-B (ViT)	58.9	MixMAE: Mixed and Masked Autoencoder for Efficient Pretraining of Hierarchical Vision Transformers

0 of 7 row(s) selected.