URIE: Universal Image Enhancementfor Visual Recognition in the Wild

Related tags

Deep Learningurie
Overview

URIE: Universal Image Enhancementfor Visual Recognition in the Wild

This is the implementation of the paper "URIE: Universal Image Enhancement for Visual Recognition in the Wild" by T. Son, J. Kang, N. Kim, S. Cho and S. Kwak. Implemented on Python 3.7 and PyTorch 1.3.1.

urie_arch

For more information, check our project website and the paper on arxiv.

Requirements

You can install dependencies using

pip install -r requirements.txt

Datasets

You need to manually configure following environment variables to run the experiments.
All validation csv contains fixed combination of image, corruption and severity to guarantee the same result.
To conduct validation, you may need to change home folder path in each csv files given.

# DATA PATHS
export IMAGENET_ROOT=PATH_TO_IMAGENET
export IMAGENET_C_ROOT=PATH_TO_IMAGENET_C

# URIE VALIDATION

## ILSVRC VALIDATION
export IMAGENET_CLN_TNG_CSV=PROJECT_PATH/imagenet_dataset/imagenet_cln_train.csv
export IMAGENET_CLN_VAL_CSV=PROJECT_PATH/imagenet_dataset/imagenet_cln_val.csv
export IMAGENET_TNG_VAL_CSV=PROJECT_PATH/imagenet_dataset/imagenet_tng_tsfrm_validation.csv
export IMAGENET_VAL_VAL_CSV=PROJECT_PATH/imagenet_dataset/imagenet_val_tsfrm_validation.csv

## CUB VALIDATION
export CUB_IMAGE=PATH_TO_CUB
export DISTORTED_CUB_IMAGE=PATH_TO_CUB_C
export CUB_TNG_LABEL=PROJECT_PATH/datasets/eval_set/label_train_cub200_2011.csv
export CUB_VAL_LABEL=PROJECT_PATH/datasets/eval_set/label_val_cub200_2011.csv
export CUB_TNG_TRAIN_VAL=PROJECT_PATH/datasets/eval_set/tng_tsfrm_validation.csv
export CUB_TNG_TEST_VAL=PROJECT_PATH/datasets/eval_set/val_tsfrm_validation.csv

ILSVRC2012 Dataset

You can download the dataset from here and use it for training.

CUB dataset

You can download the original Caltech-UCSD Birds-200-2011 dataset from here, and corrupted version of CUB dataset from here.

Training

Training URIE with the proposed method on ILSVRC2012 dataset

python train_urie.py --batch_size BATCH_SIZE \
                     --cuda \
                     --test_batch_size BATCH_SIZE \
                     --epochs 60 \
                     --lr 0.0001 \
                     --seed 5000 \
                     --desc DESCRIPTION \
                     --save SAVE_PATH \
                     --load_classifier \
                     --dataset ilsvrc \
                     --backbone r50 \
                     --multi

Since training on ILSVRC dataset takes too long, you can train / test the model with cub dataset with following command.

python train_urie.py --batch_size BATCH_SIZE \
                     --cuda \
                     --test_batch_size BATCH_SIZE \
                     --epochs 60 \
                     --lr 0.0001 \
                     --seed 5000 \
                     --desc DESCRIPTION \
                     --save SAVE_PATH \
                     --load_classifier \
                     --dataset cub \
                     --backbone r50 \
                     --multi

Validation

You may use our pretrained model to validate or compare the results.

Classification

python inference.py --srcnn_pretrained_path PROJECT_PATH/ECCV_MODELS/ECCV_SKUNET_OURS.ckpt.pt \
                    --dataset DATASET \
                    --test_batch_size 32 \
                    --enhancer ours \
                    --recog r50

Detection

We have conducted object detection experiments using the codes from github.
You may compare the performance with the same evaluation code with attaching our model (or yours) in front of the detection model.

For valid comparison, you need to preprocess your data with mean and standard deviation.

Semantic Segmentation

We have conducted semantic segmentation experiments using the codes from github.
For backbone segmentation network, please you pretrained deeplabv3 on pytorch. You may compare the performance with the same evaluation code with attaching our model (or yours) in front of the segmentation model.

For valid comparison, you need to preprocess your data with mean and standard deviation.

Image Comparison

If you want just simple before & output image comparison, you can use render.py as following command.

python render.py IMAGE_FILE_PATH

Comparison
It runs given image file through pretrained URIE model, and saves enhanced output image comparison in current project file as "output.jpg".

BibTeX

If you use this code for your research, please consider citing:

@InProceedings{son2020urie,
  title={URIE: Universal Image Enhancement for Visual Recognition in the Wild},
  author={Son, Taeyoung and Kang, Juwon and Kim, Namyup and Cho, Sunghyun and Kwak, Suha},
  booktitle={ECCV},
  year={2020}
}
Owner
Taeyoung Son
Graduate student at POSTECH, South Korea
Taeyoung Son
Change is Everywhere: Single-Temporal Supervised Object Change Detection in Remote Sensing Imagery (ICCV 2021)

Change is Everywhere Single-Temporal Supervised Object Change Detection in Remote Sensing Imagery by Zhuo Zheng, Ailong Ma, Liangpei Zhang and Yanfei

Zhuo Zheng 125 Dec 13, 2022
Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecasting

Autoformer (NeurIPS 2021) Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecasting Time series forecasting is a c

THUML @ Tsinghua University 847 Jan 08, 2023
Visualizer for neural network, deep learning, and machine learning models

Netron is a viewer for neural network, deep learning and machine learning models. Netron supports ONNX (.onnx, .pb, .pbtxt), Keras (.h5, .keras), Tens

Lutz Roeder 21k Jan 06, 2023
unofficial pytorch implement of "Squareplus: A Softplus-Like Algebraic Rectifier"

SquarePlus (Pytorch implement) unofficial pytorch implement of "Squareplus: A Softplus-Like Algebraic Rectifier" SquarePlus Squareplus is a Softplus-L

SeeFun 3 Dec 29, 2021
Like a cowsay but without cows!

Foxsay This is a simple program that generates pictures of a cute fox with a message. It is like a cowsay but without cows! Fox girls are better! Usag

Anastasia Kim 28 Feb 20, 2022
A PyTorch implementation for our paper "Dual Contrastive Learning: Text Classification via Label-Aware Data Augmentation".

Dual-Contrastive-Learning A PyTorch implementation for our paper "Dual Contrastive Learning: Text Classification via Label-Aware Data Augmentation". Y

hoshi-hiyouga 85 Dec 26, 2022
My 1st place solution at Kaggle Hotel-ID 2021

1st place solution at Kaggle Hotel-ID My 1st place solution at Kaggle Hotel-ID to Combat Human Trafficking 2021. https://www.kaggle.com/c/hotel-id-202

Kohei Ozaki 18 Aug 19, 2022
Improving adversarial robustness by a coupling rejection strategy

Adversarial Training with Rectified Rejection The code for the paper Adversarial Training with Rectified Rejection. Environment settings and libraries

Tianyu Pang 29 Jan 06, 2023
Deep Networks with Recurrent Layer Aggregation

RLA-Net: Recurrent Layer Aggregation Recurrence along Depth: Deep Networks with Recurrent Layer Aggregation This is an implementation of RLA-Net (acce

Joy Fang 21 Aug 16, 2022
Repository for the paper : Meta-FDMixup: Cross-Domain Few-Shot Learning Guided byLabeled Target Data

1 Meta-FDMIxup Repository for the paper : Meta-FDMixup: Cross-Domain Few-Shot Learning Guided byLabeled Target Data. (ACM MM 2021) paper News! the rep

Fu Yuqian 44 Nov 18, 2022
Grammar Induction using a Template Tree Approach

Gitta Gitta ("Grammar Induction using a Template Tree Approach") is a method for inducing context-free grammars. It performs particularly well on data

Thomas Winters 36 Nov 15, 2022
Official PyTorch Implementation of Convolutional Hough Matching Networks, CVPR 2021 (oral)

Convolutional Hough Matching Networks This is the implementation of the paper "Convolutional Hough Matching Network" by J. Min and M. Cho. Implemented

Juhong Min 70 Nov 22, 2022
Codes and models of NeurIPS2021 paper - DominoSearch: Find layer-wise fine-grained N:M sparse schemes from dense neural networks

DominoSearch This is repository for codes and models of NeurIPS2021 paper - DominoSearch: Find layer-wise fine-grained N:M sparse schemes from dense n

11 Sep 10, 2022
Official Repository for Machine Learning class - Physics Without Frontiers 2021

PWF 2021 Física Sin Fronteras es un proyecto del Centro Internacional de Física Teórica (ICTP) en Trieste Italia. El ICTP es un centro dedicado a fome

36 Aug 06, 2022
Open Source Light Field Toolbox for Super-Resolution

BasicLFSR BasicLFSR is an open-source and easy-to-use Light Field (LF) image Super-Ressolution (SR) toolbox based on PyTorch, including a collection o

Squidward 50 Nov 18, 2022
Using Streamlit to host a multi-page tool with model specs and classification metrics, while also accepting user input values for prediction.

Predicitng_viability Using Streamlit to host a multi-page tool with model specs and classification metrics, while also accepting user input values for

Gopalika Sharma 1 Nov 08, 2021
Unofficial implementation of Google "CutPaste: Self-Supervised Learning for Anomaly Detection and Localization" in PyTorch

CutPaste CutPaste: image from paper Unofficial implementation of Google's "CutPaste: Self-Supervised Learning for Anomaly Detection and Localization"

Lilit Yolyan 59 Nov 27, 2022
MoveNet Single Pose on OpenVINO

MoveNet Single Pose tracking on OpenVINO Running Google MoveNet Single Pose models on OpenVINO. A convolutional neural network model that runs on RGB

35 Nov 11, 2022
Sparse Progressive Distillation: Resolving Overfitting under Pretrain-and-Finetune Paradigm

Sparse Progressive Distillation: Resolving Overfitting under Pretrain-and-Finetu

3 Dec 05, 2022
Scalable Attentive Sentence-Pair Modeling via Distilled Sentence Embedding (AAAI 2020) - PyTorch Implementation

Scalable Attentive Sentence-Pair Modeling via Distilled Sentence Embedding PyTorch implementation for the Scalable Attentive Sentence-Pair Modeling vi

Microsoft 25 Dec 02, 2022