Long Tail Learning On Inaturalist 2018

평가 지표

Top-1 Accuracy

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	Top-1 Accuracy	Paper Title	Repository
LADE	70.0%	Disentangling Label Distribution for Long-tailed Visual Recognition
CB-LWS	69.5%	Decoupling Representation and Classifier for Long-Tailed Recognition
DisAlign	70.6%	Distribution Alignment: A Unified Framework for Long-tail Visual Recognition
TSC(ResNet-50)	69.7%	Targeted Supervised Contrastive Learning for Long-Tailed Recognition
GPaCo (2-R152)	79.8%	Generalized Parametric Contrastive Learning
BatchFormer(ResNet-50, RIDE)	74.1%	BatchFormer: Learning to Explore Sample Relationships for Robust Representation Learning
RIDE-DiVE	73.44%	Distilling Virtual Examples for Long-tailed Recognition
LDAM + DRW + SAM	70.1	Escaping Saddle Points for Effective Generalization on Class-Imbalanced Data
Online Feature Augmentation	65.9%	Feature Space Augmentation for Long-Tailed Data	-
cRT + SSP	68.1%	Rethinking the Value of Labels for Improving Class-Imbalanced Learning
DeiT-LT	75.1%	DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets
LIFT (ViT-B/16)	80.4%	Long-Tail Learning with Foundation Model: Heavy Fine-Tuning Hurts
MiSLAS	71.6%	Improving Calibration for Long-Tailed Recognition
LDAM-DRS-RSG	70.3%	RSG: A Simple but Effective Module for Learning Imbalanced Datasets
VL-LTR (ViT-B-16)	81.0%	VL-LTR: Learning Class-wise Visual-Linguistic Representation for Long-Tailed Visual Recognition
NCL(ResNet-50)	74.2%	Nested Collaborative Learning for Long-Tailed Visual Recognition	-
LTR-weight-balancing(ResNet-50)	70.2%	Long-Tailed Recognition via Weight Balancing
Hybrid-PSC	68.1%	Contrastive Learning based Hybrid Networks for Long-Tailed Image Classification	-
CE-DRW-IC	67.9%	Posterior Re-calibration for Imbalanced Datasets	-
PaCo(ResNet-152)	75.2%	Parametric Contrastive Learning

0 of 43 row(s) selected.