Image Classification On Objectnet

评估指标

Top-1 Accuracy

评测结果

各个模型在此基准测试上的表现结果

		Paper Title	Repository
CoCa	82.7	CoCa: Contrastive Captioners are Image-Text Foundation Models
LiT	82.5	LiT: Zero-Shot Transfer with Locked-image text Tuning
BASIC	82.3	Combined Scaling for Zero-shot Transfer Learning	-
EVA-02-CLIP-E/14+	79.6	EVA-CLIP: Improved Training Techniques for CLIP at Scale
Baseline (ViT-G/14)	79.03	Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time
Model soups (ViT-G/14)	78.52	Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time
MAWS (ViT-6.5B)	77.9	The effectiveness of MAE pre-pretraining for billion-scale pretraining
MAWS (ViT-2B)	75.8	The effectiveness of MAE pre-pretraining for billion-scale pretraining
MAWS (ViT-H)	72.6	The effectiveness of MAE pre-pretraining for billion-scale pretraining
CLIP	72.3	Learning Transferable Visual Models From Natural Language Supervision
ALIGN	72.2	Combined Scaling for Zero-shot Transfer Learning	-
WiSE-FT	72.1	Robust fine-tuning of zero-shot models
ViT-e	72.0	PaLI: A Jointly-Scaled Multilingual Language-Image Model
ViT-G/14	70.53	Scaling Vision Transformers
SWAG (ViT H/14)	69.5	Revisiting Weakly Supervised Pre-Training of Visual Perception Models
NS (Eff.-L2)	68.5	Scaling Vision Transformers
RegNetY 128GF (Platt)	64.3	Revisiting Weakly Supervised Pre-Training of Visual Perception Models
LLE (ViT-H/14, MAE, Edge Aug)	60.78	A Whac-A-Mole Dilemma: Shortcuts Come in Multiples Where Mitigating One Amplifies Others
SEER (RegNet10B)	60.2	Vision Models Are More Robust And Fair When Pretrained On Uncurated Images Without Supervision
ViT H/14 (Platt)	60	Revisiting Weakly Supervised Pre-Training of Visual Perception Models

0 of 106 row(s) selected.