Image Classification on Caltech-101

Metric: Accuracy (higher is better)

LeaderboardDataset

Loading chart...

Results

Submit a result

Hide extra data

Sort:

#	Model↕	Accuracy▼	Extra Data	Paper	Date↕	Code
1	Pre trained wide-resnet-101	97.76	No	ProgressiveSpinalNet architecture for FC layers	2021-03-21	Code
2	Wide-ResNet-101 (Spinal FC)	97.32	Yes	SpinalNet: Deep Neural Network with Gradual Input	2020-07-07	Code
3	ResNeXt-101-32x8d	95.58	No	Dead Pixel Test Using Effective Receptive Field	2021-08-31	Code
4	Bamboo (ViT-B/16)	94.8	Yes	Bamboo: Building Mega-Scale Vision Dataset Conti...	2022-03-15	Code
5	SEER (RegNet10B - linear eval)	91	Yes	Vision Models Are More Robust And Fair When Pret...	2022-02-16	Code
6	UL-Hopfield (ULH)	91	No	Unsupervised Learning using Pretrained CNN and A...	2018-05-02	-

#1Pre trained wide-resnet-101SOTA
97.76
Accuracy· 2021-03-21
ProgressiveSpinalNet architecture for FC layers Code
#2Wide-ResNet-101 (Spinal FC)SOTA
97.32
Accuracy· Extra Data· 2020-07-07
SpinalNet: Deep Neural Network with Gradual Input Code
#3ResNeXt-101-32x8d
95.58
Accuracy· 2021-08-31
Dead Pixel Test Using Effective Receptive Field Code
#4Bamboo (ViT-B/16)
94.8
Accuracy· Extra Data· 2022-03-15
Bamboo: Building Mega-Scale Vision Dataset Continually with Human-Machine Synergy Code
#5SEER (RegNet10B - linear eval)
91
Accuracy· Extra Data· 2022-02-16
Vision Models Are More Robust And Fair When Pretrained On Uncurated Images Without Supervision Code
#6UL-Hopfield (ULH)SOTA
91
Accuracy· 2018-05-02
Unsupervised Learning using Pretrained CNN and Associative Memory Bank