SCAN: Learning to Classify Images without Labels, incl. SimCLR. [ECCV 2020]

Overview

Learning to Classify Images without Labels

This repo contains the Pytorch implementation of our paper:

SCAN: Learning to Classify Images without Labels

Wouter Van Gansbeke, Simon Vandenhende, Stamatios Georgoulis, Marc Proesmans and Luc Van Gool.

  • Accepted at ECCV 2020 (Slides). Watch the explanation of our paper by Yannic Kilcher on YouTube.

  • 🏆 SOTA on 4 benchmarks. Check out Papers With Code for Image Clustering or Unsup. Classification.

  • 🆕 Interested in representation learning on non-curated datasets? Check out our NeurIPS'21 paper and code.

  • 🆕 Interested in unsupervised semantic segmentation? Check out our ICCV'21 paper: MaskContrast.

  • 📜 Looking for influential papers in self-supervised learning? Check out this reading list.

PWC PWC PWC PWC

Contents

  1. Introduction
  2. Prior Work
  3. Installation
  4. Training
  5. Model Zoo
  6. Tutorial
  7. Citation

🆕 Tutorial section has been added, checkout TUTORIAL.md.

🆕 Prior work section has been added, checkout Prior Work.

Introduction

Can we automatically group images into semantically meaningful clusters when ground-truth annotations are absent? The task of unsupervised image classification remains an important, and open challenge in computer vision. Several recent approaches have tried to tackle this problem in an end-to-end fashion. In this paper, we deviate from recent works, and advocate a two-step approach where feature learning and clustering are decoupled.

We outperform state-of-the-art methods by large margins, in particular +26.6% on CIFAR10, +25.0% on CIFAR100-20 and +21.3% on STL10 in terms of classification accuracy. Our method is the first to perform well on ImageNet (1000 classes). Check out the benchmarks on the Papers-with-code website for Image Clustering and Unsupervised Image Classification.

Prior Work

  • Train set/test set: We would like to point out that most prior work in unsupervised classification use both the train and test set during training. We believe this is bad practice and therefore propose to only train on the train set. The final numbers should be reported on the test set (see table 3 of our paper). This also allows us to directly compare with supervised and semi-supervised methods in the literature. We encourage future work to do the same. We observe around 2% improvement over the reported numbers when including the test set.

  • Reproducibility: We noticed that prior work is very initialization sensitive. So, we don't think reporting a single number is therefore fair. We report our results as the mean and standard deviation over 10 runs.

Please follow the instructions underneath to perform semantic clustering with SCAN.

Installation

The code runs with recent Pytorch versions, e.g. 1.4. Assuming Anaconda, the most important packages can be installed as:

conda install pytorch=1.4.0 torchvision=0.5.0 cudatoolkit=10.0 -c pytorch
conda install matplotlib scipy scikit-learn   # For evaluation and confusion matrix visualization
conda install faiss-gpu                       # For efficient nearest neighbors search 
conda install pyyaml easydict                 # For using config files
conda install termcolor                       # For colored print statements

We refer to the requirements.txt file for an overview of the packages in the environment we used to produce our results.

Training

Setup

The following files need to be adapted in order to run the code on your own machine:

  • Change the file paths to the datasets in utils/mypath.py, e.g. /path/to/cifar10.
  • Specify the output directory in configs/env.yml. All results will be stored under this directory.

Our experimental evaluation includes the following datasets: CIFAR10, CIFAR100-20, STL10 and ImageNet. The ImageNet dataset should be downloaded separately and saved to the path described in utils/mypath.py. Other datasets will be downloaded automatically and saved to the correct path when missing.

Train model

The configuration files can be found in the configs/ directory. The training procedure consists of the following steps:

  • STEP 1: Solve the pretext task i.e. simclr.py
  • STEP 2: Perform the clustering step i.e. scan.py
  • STEP 3: Perform the self-labeling step i.e. selflabel.py

For example, run the following commands sequentially to perform our method on CIFAR10:

python simclr.py --config_env configs/your_env.yml --config_exp configs/pretext/simclr_cifar10.yml
python scan.py --config_env configs/your_env.yml --config_exp configs/scan/scan_cifar10.yml
python selflabel.py --config_env configs/your_env.yml --config_exp configs/selflabel/selflabel_cifar10.yml

Remarks

The provided hyperparameters are identical for CIFAR10, CIFAR100-20 and STL10. However, fine-tuning the hyperparameters can further improve the results. We list the most important hyperparameters of our method below:

  • Entropy weight: Can be adapted when the number of clusters changes. In general, try to avoid imbalanced clusters during training.
  • Confidence threshold: When every cluster contains a sufficiently large amount of confident samples, it can be beneficial to increase the threshold. This generally helps to decrease the noise. The ablation can be found in the paper.
  • Number of neighbors in SCAN: The dependency on this hyperparameter is rather small as shown in the paper.

Model Zoo

Pretext tasks

We perform the instance discrimination task in accordance with the scheme from SimCLR on CIFAR10, CIFAR100 and STL10. Pretrained models can be downloaded from the links listed below. On ImageNet, we use the pretrained weights provided by MoCo and transfer them to be compatible with our code repository.

Dataset Download link
CIFAR10 Download
CIFAR100 Download
STL10 Download

Clustering

We provide the following pretrained models after training with the SCAN-loss, and after the self-labeling step. The best models can be found here and we futher refer to the paper for the averages and standard deviations.

Dataset Step ACC NMI ARI Download link
CIFAR10 SCAN-loss 81.6 71.5 66.5 Download
Self-labeling 88.3 79.7 77.2 Download
CIFAR100 SCAN-loss 44.0 44.9 28.3 Download
Self-labeling 50.7 48.6 33.3 Download
STL10 SCAN-loss 79.2 67.3 61.8 Download
Self-labeling 80.9 69.8 64.6 Download
ImageNet-50 SCAN-loss 75.1 80.5 63.5 Download
Self-labeling 76.8 82.2 66.1 Download
ImageNet-100 SCAN-loss 66.2 78.7 54.4 Download
Self-labeling 68.9 80.8 57.6 Download
ImageNet-200 SCAN-loss 56.3 75.7 44.1 Download
Self-labeling 58.1 77.2 47.0 Download

Result ImageNet

We also train SCAN on ImageNet for 1000 clusters. We use 10 clusterheads and finally take the head with the lowest loss. The accuracy (ACC), normalized mutual information (NMI), adjusted mutual information (AMI) and adjusted rand index (ARI) are computed:

Method ACC NMI AMI ARI Download link
SCAN (ResNet50) 39.9 72.0 51.2 27.5 Download

Evaluation

Pretrained models from the model zoo can be evaluated using the eval.py script. For example, the model on cifar-10 can be evaluated as follows:

python eval.py --config_exp configs/scan/scan_cifar10.yml --model $MODEL_PATH 

Visualizing the prototype images is easily done by setting the --visualize_prototypes flag. For example on cifar-10:

Similarly, you might want to have a look at the clusters found on ImageNet (as shown at the top). First download the model (link in table above) and then execute the following command:

python eval.py --config_exp configs/scan/imagenet_eval.yml --model $MODEL_PATH_IMAGENET 

Tutorial

If you want to see another (more detailed) example for STL-10, checkout TUTORIAL.md. It provides a detailed guide and includes visualizations and log files with the training progress.

Citation

If you find this repo useful for your research, please consider citing our paper:

@inproceedings{vangansbeke2020scan,
  title={Scan: Learning to classify images without labels},
  author={Van Gansbeke, Wouter and Vandenhende, Simon and Georgoulis, Stamatios and Proesmans, Marc and Van Gool, Luc},
  booktitle={Proceedings of the European Conference on Computer Vision},
  year={2020}
}

For any enquiries, please contact the main authors.

License

This software is released under a creative commons license which allows for personal and research use only. For a commercial license please contact the authors. You can view a license summary here.

Acknoledgements

This work was supported by Toyota, and was carried out at the TRACE Lab at KU Leuven (Toyota Research on Automated Cars in Europe - Leuven).

Owner
Wouter Van Gansbeke
PhD researcher at KU Leuven. Especially interested in computer vision, machine learning and deep learning. Working on self-supervised and multi-task learning.
Wouter Van Gansbeke
Bayesian Meta-Learning Through Variational Gaussian Processes

vmgp This is the repository of Vivek Myers and Nikhil Sardana for our CS 330 final project, Bayesian Meta-Learning Through Variational Gaussian Proces

Vivek Myers 2 Nov 17, 2022
A Pytorch implementation of "LegoNet: Efficient Convolutional Neural Networks with Lego Filters" (ICML 2019).

LegoNet This code is the implementation of ICML2019 paper LegoNet: Efficient Convolutional Neural Networks with Lego Filters Run python train.py You c

YangZhaohui 140 Sep 26, 2022
Deep learning library for solving differential equations and more

DeepXDE Voting on whether we should have a Slack channel for discussion. DeepXDE is a library for scientific machine learning. Use DeepXDE if you need

Lu Lu 1.4k Dec 29, 2022
This code reproduces the results of the paper, "Measuring Data Leakage in Machine-Learning Models with Fisher Information"

Fisher Information Loss This repository contains code that can be used to reproduce the experimental results presented in the paper: Awni Hannun, Chua

Facebook Research 43 Dec 30, 2022
Dahua Camera and Doorbell Home Assistant Integration

Home Assistant Dahua Integration The Dahua Home Assistant integration allows you to integrate your Dahua cameras and doorbells in Home Assistant. It's

Ronnie 216 Dec 26, 2022
Capstone-Project-2 - A game program written in the Python language

Capstone-Project-2 My Pygame Game Information: Description This Pygame project i

Nhlakanipho Khulekani Hlophe 1 Jan 04, 2022
Training deep models using anime, illustration images.

animeface deep models for anime images. Datasets anime-face-dataset Anime faces collected from Getchu.com. Based on Mckinsey666's dataset. 63.6K image

Tomoya Sawada 61 Dec 25, 2022
A Closer Look at Structured Pruning for Neural Network Compression

A Closer Look at Structured Pruning for Neural Network Compression Code used to reproduce experiments in https://arxiv.org/abs/1810.04622. To prune, w

Bayesian and Neural Systems Group 140 Dec 05, 2022
Sarus implementation of classical ML models. The models are implemented using the Keras API of tensorflow 2. Vizualization are implemented and can be seen in tensorboard.

Sarus published models Sarus implementation of classical ML models. The models are implemented using the Keras API of tensorflow 2. Vizualization are

Sarus Technologies 39 Aug 19, 2022
A tool for making map images from OpenTTD save games

OpenTTD Surveyor A tool for making map images from OpenTTD save games. This is not part of the main OpenTTD codebase, nor is it ever intended to be pa

Aidan Randle-Conde 9 Feb 15, 2022
A PyTorch re-implementation of the paper 'Exploring Simple Siamese Representation Learning'. Reproduced the 67.8% Top1 Acc on ImageNet.

Exploring simple siamese representation learning This is a PyTorch re-implementation of the SimSiam paper on ImageNet dataset. The results match that

Taojiannan Yang 72 Nov 09, 2022
Implementation of DropLoss for Long-Tail Instance Segmentation in Pytorch

[AAAI 2021]DropLoss for Long-Tail Instance Segmentation [AAAI 2021] DropLoss for Long-Tail Instance Segmentation Ting-I Hsieh*, Esther Robb*, Hwann-Tz

Tim 37 Dec 02, 2022
Official implementation of "SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers"

SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers Figure 1: Performance of SegFormer-B0 to SegFormer-B5. Project page

NVIDIA Research Projects 1.4k Dec 31, 2022
Compact Bidirectional Transformer for Image Captioning

Compact Bidirectional Transformer for Image Captioning Requirements Python 3.8 Pytorch 1.6 lmdb h5py tensorboardX Prepare Data Please use git clone --

YE Zhou 19 Dec 12, 2022
Aesara is a Python library that allows one to define, optimize, and efficiently evaluate mathematical expressions involving multi-dimensional arrays.

Aesara is a Python library that allows one to define, optimize, and efficiently evaluate mathematical expressions involving multi-dimensional arrays.

Aesara 898 Jan 07, 2023
[ WSDM '22 ] On Sampling Collaborative Filtering Datasets

On Sampling Collaborative Filtering Datasets This repository contains the implementation of many popular sampling strategies, along with various expli

Noveen Sachdeva 17 Dec 08, 2022
[ ICCV 2021 Oral ] Our method can estimate camera poses and neural radiance fields jointly when the cameras are initialized at random poses in complex scenarios (outside-in scenes, even with less texture or intense noise )

GNeRF This repository contains official code for the ICCV 2021 paper: GNeRF: GAN-based Neural Radiance Field without Posed Camera. This implementation

Quan Meng 191 Dec 26, 2022
Audio Source Separation is the process of separating a mixture into isolated sounds from individual sources

Audio Source Separation is the process of separating a mixture into isolated sounds from individual sources (e.g. just the lead vocals).

Victor Basu 14 Nov 07, 2022
Code for Multinomial Diffusion

Code for Multinomial Diffusion Abstract Generative flows and diffusion models have been predominantly trained on ordinal data, for example natural ima

104 Jan 04, 2023
This repo includes the CUB-GHA (Gaze-based Human Attention) dataset and code of the paper "Human Attention in Fine-grained Classification".

HA-in-Fine-Grained-Classification This repo includes the CUB-GHA (Gaze-based Human Attention) dataset and code of the paper "Human Attention in Fine-g

16 Oct 29, 2022