1st place solution for SIIM-FISABIO-RSNA COVID-19 Detection Challenge

Overview

SIIM-COVID19-Detection

Alt text

Source code of the 1st place solution for SIIM-FISABIO-RSNA COVID-19 Detection Challenge.

1.INSTALLATION

  • Ubuntu 18.04.5 LTS
  • CUDA 10.2
  • Python 3.7.9
  • python packages are detailed separately in requirements.txt
$ conda create -n envs python=3.7.9
$ conda activate envs
$ conda install -c conda-forge gdcm
$ pip install -r requirements.txt
$ pip install git+https://github.com/ildoonet/pytorch-gradual-warmup-lr.git

2.DATASET

2.1 SIIM COVID 19 DATASET

  • download competition dataset at link then extract to ./dataset/siim-covid19-detection
$ cd src/prepare
$ python dicom2image_siim.py
$ python kfold_split.py
$ prepare_siim_annotation.py                        # effdet and yolo format
$ cp -r ../../dataset/siim-covid19-detection/images ../../dataset/lung_crop/.
$ python prepare_siim_lung_crop_annotation.py

2.2 EXTERNAL DATASET

  • download pneumothorax dataset at link then extract to ./dataset/external_dataset/pneumothorax/dicoms
  • download pneumonia dataset at link then extract to ./dataset/external_dataset/rsna-pneumonia-detection-challenge/dicoms
  • download vinbigdata dataset at link then extract to ./dataset/external_dataset/vinbigdata/dicoms
  • download chest14 dataset at link then extract to ./dataset/external_dataset/chest14/images
  • download chexpert high-resolution dataset at link or then extract to ./dataset/external_dataset/chexpert/train
  • download padchest dataset at link or then extract to ./dataset/external_dataset/padchest/images
    most of the images in bimcv and ricord overlap with siim covid trainset and testset. To avoid data-leak when training, I didn't use them. You can use script
$ cd src/prepare
$ python dicom2image_pneumothorax.py
$ python dicom2image_pneumonia.py
$ python prepare_pneumonia_annotation.py      # effdet and yolo format
$ python dicom2image_vinbigdata.py
$ python prepare_vinbigdata.py
$ python refine_data.py                       # remove unused file in chexpert + chest14 + padchest dataset
$ python resize_padchest_pneumothorax.py

dataset structure should be ./dataset/dataset_structure.txt

3.SOLUTION SUMMARY

Alt text

4.TRAIN MODEL

4.1 Classification

4.1.1 Multi head classification + segmentation

  • Stage1
$ cd src/classification_aux
$ bash train_chexpert_chest14.sh              #Pretrain backbone on chexpert + chest14
$ bash train_rsnapneu.sh                      #Pretrain rsna_pneumonia
$ bash train_siim.sh                          #Train siim covid19
  • Stage2: Generate soft-label for classification head and mask for segmentation head.
    Output: soft-label in ./pseudo_csv/[source].csv and public test masks in ./prediction_mask/public_test/masks
$ bash generate_pseudo_label.sh [checkpoints_dir]
  • Stage3: Train model on trainset + public testset, load checkpoint from previous round
$ bash train_pseudo.sh [previous_checkpoints_dir] [new_checkpoints_dir]

Rounds of pseudo labeling (stage2) and retraining (stage3) were repeated until the score on public LB didn't improve.

  • For final submission
$ bash generate_pseudo_label.sh checkpoints_v3
$ bash train_pseudo.sh checkpoints_v3 checkpoints_v4
  • For evaluation
$ CUDA_VISIBLE_DEVICES=0 python evaluate.py --cfg configs/xxx.yaml --num_tta xxx

[email protected] 4 classes: negative, typical, indeterminate, atypical

SeR152-Unet EB5-Deeplab EB6-Linknet EB7-Unet++ Ensemble
w/o TTA/8TTA 0.575/0.584 0.583/0.592 0.580/0.587 0.589/0.595 0.595/0.598

8TTA: (orig, center-crop 80%)x(None, hflip, vflip, hflip & vflip). In final submission, I use 4.1.2 lung detector instead of center-crop 80%

4.1.2 Lung Detector-YoloV5

I annotated the train data(6334 images) using LabelImg and built a lung localizer. I noticed that increasing input image size improves the modeling performance and lung detector helps the model to reduce background noise.

$ cd src/detection_lung_yolov5
$ cd weights && bash download_coco_weights.sh && cd ..
$ bash train.sh
Fold0 Fold1 Fold2 Fold3 Fold4 Average
[email protected]:0.95 0.921 0.931 0.926 0.923 0.922 0.9246
[email protected] 0.997 0.998 0.997 0.996 0.998 0.9972

4.2 Opacity Detection

Rounds of pseudo labeling (stage2) and retraining (stage3) were repeated until the score on public LB didn't improve.

4.2.1 YoloV5x6 768

  • Stage1:
$ cd src/detection_yolov5
$ cd weights && bash download_coco_weights.sh && cd ..
$ bash train_rsnapneu.sh          #pretrain with rsna_pneumonia
$ bash train_siim.sh              #train with siim covid19 dataset, load rsna_pneumonia checkpoint
  • Stage2: Generate pseudo label (boxes)
$ bash generate_pseudo_label.sh

Jump to step 4.2.4 Ensembling + Pseudo labeling

  • Stage3:
$ bash warmup_ext_dataset.sh      #train with pseudo labeling (public-test, padchest, pneumothorax, vin) + rsna_pneumonia
$ bash train_final.sh             #train siim covid19 boxes, load warmup checkpoint

4.2.2 EfficientDet D7 768

  • Stage1:
$ cd src/detection_efffdet
$ bash train_rsnapneu.sh          #pretrain with rsna_pneumonia
$ bash train_siim.sh              #train with siim covid19 dataset, load rsna_pneumonia checkpoint
  • Stage2: Generate pseudo label (boxes)
$ bash generate_pseudo_label.sh

Jump to step 4.2.4 Ensembling + Pseudo labeling

  • Stage3:
$ bash warmup_ext_dataset.sh      #train with pseudo labeling (public-test, padchest, pneumothorax, vin) + rsna_pneumonia
$ bash train_final.sh             #train siim covid19, load warmup checkpoint

4.2.3 FasterRCNN FPN 768 & 1024

  • Stage1: train backbone of model with chexpert + chest14 -> train model with rsna pneummonia -> train model with siim, load rsna pneumonia checkpoint
$ cd src/detection_fasterrcnn
$ CUDA_VISIBLE_DEVICES=0,1,2,3 python train_chexpert_chest14.py --steps 0 1 --cfg configs/resnet200d.yaml
$ CUDA_VISIBLE_DEVICES=0,1,2,3 python train_chexpert_chest14.py --steps 0 1 --cfg configs/resnet101d.yaml
$ CUDA_VISIBLE_DEVICES=0 python train_rsnapneu.py --cfg configs/resnet200d.yaml
$ CUDA_VISIBLE_DEVICES=0 python train_rsnapneu.py --cfg configs/resnet101d.yaml
$ CUDA_VISIBLE_DEVICES=0 python train_siim.py --cfg configs/resnet200d.yaml --folds 0 1 2 3 4 --SEED 123
$ CUDA_VISIBLE_DEVICES=0 python train_siim.py --cfg configs/resnet101d.yaml --folds 0 1 2 3 4 --SEED 123

Note: Change SEED if training script runs into issue related to augmentation (boundingbox area=0) and comment/uncomment the following code if training script runs into issue related to resource limit

import resource
rlimit = resource.getrlimit(resource.RLIMIT_NOFILE)
resource.setrlimit(resource.RLIMIT_NOFILE, (8192, rlimit[1]))
  • Stage2: Generate pseudo label (boxes)
$ bash generate_pseudo_label.sh

Jump to step 4.2.4 Ensembling + Pseudo labeling

  • Stage3:
$ CUDA_VISIBLE_DEVICES=0 python warmup_ext_dataset.py --cfg configs/resnet200d.yaml
$ CUDA_VISIBLE_DEVICES=0 python warmup_ext_dataset.py --cfg configs/resnet101d.yaml
$ CUDA_VISIBLE_DEVICES=0 python train_final.py --cfg configs/resnet200d.yaml
$ CUDA_VISIBLE_DEVICES=0 python train_final.py --cfg configs/resnet101d.yaml

4.2.4 Ensembling + Pseudo labeling

Keep images that meet the conditions: negative prediction < 0.3 and maximum of (typical, indeterminate, atypical) predicion > 0.7. Then choose 2 boxes with the highest confidence as pseudo labels for each image.

Note: This step requires at least 128 GB of RAM

$ cd ./src/detection_make_pseudo
$ python make_pseudo.py
$ python make_annotation.py            

4.2.5 Detection Performance

YoloV5x6 768 EffdetD7 768 F-RCNN R200 768 F-RCNN R101 1024
[email protected] TTA 0.580 0.594 0.592 0.596

Final result: Public LB/Private LB: 0.658/0.635

5.FINAL KERNEL

siim-covid19-2021
demo notebook to visualize output of models

6.AWESOME RESOURCES

Pytorch
PyTorch Image Models
Segmentation models
EfficientDet
YoloV5
FasterRCNN FPN
Albumentations
Weighted boxes fusion

Owner
Nguyen Ba Dung
https://www.linkedin.com/in/dungnb1333/
Nguyen Ba Dung
Responsive Doc. scanner using U^2-Net, Textcleaner and Tesseract

Responsive Doc. scanner using U^2-Net, Textcleaner and Tesseract Toolset U^2-Net is used for background removal Textcleaner is used for image cleaning

3 Jul 13, 2022
A list of hyperspectral image super-solution resources collected by Junjun Jiang

A list of hyperspectral image super-resolution resources collected by Junjun Jiang. If you find that important resources are not included, please feel free to contact me.

Junjun Jiang 301 Jan 05, 2023
OpenCV-Erlang/Elixir bindings

evision [WIP] : OS : arch Build Status Ubuntu 20.04 arm64 Ubuntu 20.04 armv7 Ubuntu 20.04 s390x Ubuntu 20.04 ppc64le Ubuntu 20.04 x86_64 macOS 11 Big

Cocoa 194 Jan 05, 2023
EAST for ICPR MTWI 2018 Challenge II (Text detection of network images)

EAST_ICPR2018: EAST for ICPR MTWI 2018 Challenge II (Text detection of network images) Introduction This is a repository forked from argman/EAST for t

QichaoWu 49 Dec 24, 2022
An organized collection of tutorials and projects created for aspriring computer vision students.

A repository created with the purpose of teaching students in BME lab 308A- Hanoi University of Science and Technology

Givralnguyen 5 Nov 24, 2021
Code for the "Sensing leg movement enhances wearable monitoring of energy expenditure" paper.

EnergyExpenditure Code for the "Sensing leg movement enhances wearable monitoring of energy expenditure" paper. Additional data for replicating this s

Patrick S 42 Oct 26, 2022
Random maze generator and solver

Maze Generator and Solver I wrote a maze generator that works with two commonly known algorithms: Depth First Search and Randomized Prims. Both of the

Daniel Pérez 10 Sep 23, 2022
A curated list of resources dedicated to scene text localization and recognition

Scene Text Localization & Recognition Resources A curated list of resources dedicated to scene text localization and recognition. Any suggestions and

CarlosTao 1.6k Dec 22, 2022
Scan the MRZ code of a passport and extract the firstname, lastname, passport number, nationality, date of birth, expiration date and personal numer.

PassportScanner Works with 2 and 3 line identity documents. What is this With PassportScanner you can use your camera to scan the MRZ code of a passpo

Edwin Vermeer 441 Dec 24, 2022
Primary QPDF source code and documentation

QPDF QPDF is a command-line tool and C++ library that performs content-preserving transformations on PDF files. It supports linearization, encryption,

QPDF 2.2k Jan 04, 2023
An application of high resolution GANs to dewarp images of perturbed documents

Docuwarp This project is focused on dewarping document images through the usage of pix2pixHD, a GAN that is useful for general image to image translat

Thomas Huang 97 Dec 25, 2022
OCR, Scene-Text-Understanding, Text Recognition

Scene-Text-Understanding Survey [2015-PAMI] Text Detection and Recognition in Imagery: A Survey paper [2014-Front.Comput.Sci] Scene Text Detection and

Alan Tang 354 Dec 12, 2022
Code release for our paper, "SimNet: Enabling Robust Unknown Object Manipulation from Pure Synthetic Data via Stereo"

SimNet: Enabling Robust Unknown Object Manipulation from Pure Synthetic Data via Stereo Thomas Kollar, Michael Laskey, Kevin Stone, Brijen Thananjeyan

68 Dec 14, 2022
Convolutional Recurrent Neural Network (CRNN) for image-based sequence recognition.

Convolutional Recurrent Neural Network This software implements the Convolutional Recurrent Neural Network (CRNN), a combination of CNN, RNN and CTC l

Baoguang Shi 2k Dec 31, 2022
The first open-source library that detects the font of a text in a image.

Typefont Typefont is an experimental library that detects the font of a text in a image. Usage Import the main function and invoke it like in the foll

Vasile Pește 1.6k Feb 24, 2022
When Age-Invariant Face Recognition Meets Face Age Synthesis: A Multi-Task Learning Framework (CVPR 2021 oral)

MTLFace This repository contains the PyTorch implementation and the dataset of the paper: When Age-Invariant Face Recognition Meets Face Age Synthesis

Hzzone 120 Jan 05, 2023
Table recognition inside douments using neural networks

TableTrainNet A simple project for training and testing table recognition in documents. This project was developed to make a neural network which reco

Giovanni Cavallin 93 Jul 24, 2022
Characterizing possible failure modes in physics-informed neural networks.

Characterizing possible failure modes in physics-informed neural networks This repository contains the PyTorch source code for the experiments in the

Aditi Krishnapriyan 55 Jan 02, 2023
The code of "Mask TextSpotter: An End-to-End Trainable Neural Network for Spotting Text with Arbitrary Shapes"

Mask TextSpotter A Pytorch implementation of Mask TextSpotter along with its extension can be find here Introduction This is the official implementati

Pengyuan Lyu 261 Nov 21, 2022
This project modify tensorflow object detection api code to predict oriented bounding boxes. It can be used for scene text detection.

This is an oriented object detector based on tensorflow object detection API. Most of the code is not changed except for those related to the need of

Dafang He 30 Oct 22, 2022