Parametric-Contrastive-Learning

This repository contains the implementation code for ICCV2021 paper:
Parametric Contrastive Learning (https://arxiv.org/abs/2107.12028)

If you find this code or idea useful, please consider citing our work:

@misc{cui2021parametric,
      title={Parametric Contrastive Learning}, 
      author={Jiequan Cui and Zhisheng Zhong and Shu Liu and Bei Yu and Jiaya Jia},
      year={2021},
      eprint={2107.12028},
      archivePrefix={arXiv},
      primaryClass={cs.CV}
}

Overview

In this paper, we propose Parametric Contrastive Learning (PaCo) to tackle long-tailed recognition. Based on theoretical analysis, we observe supervised contrastive loss tends to bias on high-frequency classes and thus increases the difficulty of imbalance learning. We introduce a set of parametric class-wise learnable centers to rebalance from an optimization perspective. Further, we analyze our PaCo loss under a balanced setting. Our analysis demonstrates that PaCo can adaptively enhance the intensity of pushing samples of the same class close as more samples are pulled together with their corresponding centers and benefit hard example learning. Experiments on long-tailed CIFAR, ImageNet, Places, and iNaturalist 2018 manifest the new state-of-the-art for long-tailed recognition. On full ImageNet, models trained with PaCo loss surpass supervised contrastive learning across various ResNet backbones.

Results and Pretrained models

Full ImageNet (Balanced setting)

Method	Model	Top-1 Acc(%)	link	log
PaCo	ResNet-50	79.3	download	download
PaCo	ResNet-101	80.9	download	download
PaCo	ResNet-200	81.8	download	download

ImageNet-LT (Imbalance setting)

Method	Model	Top-1 Acc(%)	link	log
PaCo	ResNet-50	57.0	download	download
PaCo	ResNeXt-50	58.2	download	download
PaCo	ResNeXt-101	60.0	download	download

iNaturalist 2018 (Imbalanced setting)

Method	Model	Top-1 Acc(%)	link	log
PaCo	ResNet-50	73.2	download	download
PaCo	ResNet-152	75.2	download	download

Places-LT (Imbalanced setting)

Method	Model	Top-1 Acc(%)	link	log
PaCo	ResNet-152	41.2	download	download

Get Started

For full ImageNet, ImageNet-LT, iNaturalist 2018, Places-LT training and evaluation. Note that PyTorch>=1.6. All experiments are conducted on 4 GPUs. If you have more GPU resources, please make sure that the learning rate should be linearly scaled and 32 images per gpu is recommented.

cd Full-ImageNet
bash sh/train_resnet50.sh
bash sh/eval_resnet50.sh

cd LT
bash sh/ImageNetLT_train_R50.sh
bash sh/ImageNetLT_eval_R50.sh
bash sh/PlacesLT_train_R152.sh
bash sh/PlacesLT_eval_R152.sh