a514514772 / Dise Domain Invariant Structure Extraction
Licence: gpl-3.0
Pytorch Implementation -- All about Structure: Adapting Structural Information across Domains for Boosting Semantic Segmentation, CVPR 2019
Stars: ✭ 129
Programming Languages
python
139335 projects - #7 most used programming language
Projects that are alternatives of or similar to Dise Domain Invariant Structure Extraction
Seg Uncertainty
IJCAI2020 & IJCV 2020 🌇 Unsupervised Scene Adaptation with Memory Regularization in vivo
Stars: ✭ 202 (+56.59%)
Mutual labels: semantic-segmentation, domain-adaptation
Clan
( CVPR2019 Oral ) Taking A Closer Look at Domain Shift: Category-level Adversaries for Semantics Consistent Domain Adaptation
Stars: ✭ 248 (+92.25%)
Mutual labels: semantic-segmentation, domain-adaptation
Cbst
Code for <Domain Adaptation for Semantic Segmentation via Class-Balanced Self-Training> in ECCV18
Stars: ✭ 146 (+13.18%)
Mutual labels: semantic-segmentation, domain-adaptation
multichannel-semseg-with-uda
Multichannel Semantic Segmentation with Unsupervised Domain Adaptation
Stars: ✭ 19 (-85.27%)
Mutual labels: semantic-segmentation, domain-adaptation
LoveDA
[NeurIPS2021 Poster] LoveDA: A Remote Sensing Land-Cover Dataset for Domain Adaptive Semantic Segmentation
Stars: ✭ 111 (-13.95%)
Mutual labels: semantic-segmentation, domain-adaptation
Seanet
Self-Ensembling Attention Networks: Addressing Domain Shift for Semantic Segmentation
Stars: ✭ 90 (-30.23%)
Mutual labels: semantic-segmentation, domain-adaptation
Intrada
Unsupervised Intra-domain Adaptation for Semantic Segmentation through Self-Supervision (CVPR 2020 Oral)
Stars: ✭ 211 (+63.57%)
Mutual labels: semantic-segmentation, domain-adaptation
IAST-ECCV2020
IAST: Instance Adaptive Self-training for Unsupervised Domain Adaptation (ECCV 2020) https://teacher.bupt.edu.cn/zhuchuang/en/index.htm
Stars: ✭ 84 (-34.88%)
Mutual labels: semantic-segmentation, domain-adaptation
AdaptationSeg
Curriculum Domain Adaptation for Semantic Segmentation of Urban Scenes, ICCV 2017
Stars: ✭ 128 (-0.78%)
Mutual labels: semantic-segmentation, domain-adaptation
Adaptsegnet
Learning to Adapt Structured Output Space for Semantic Segmentation, CVPR 2018 (spotlight)
Stars: ✭ 654 (+406.98%)
Mutual labels: semantic-segmentation, domain-adaptation
Lsd Seg
Learning from Synthetic Data: Addressing Domain Shift for Semantic Segmentation
Stars: ✭ 99 (-23.26%)
Mutual labels: semantic-segmentation, domain-adaptation
Openvehiclevision
An opensource lib. for vehicle vision applications (written by MATLAB), lane marking detection, road segmentation
Stars: ✭ 120 (-6.98%)
Mutual labels: semantic-segmentation
Opencompounddomainadaptation Ocda
Pytorch implementation for "Open Compound Domain Adaptation" (CVPR 2020 ORAL)
Stars: ✭ 114 (-11.63%)
Mutual labels: domain-adaptation
Cen
[NeurIPS 2020] Code release for paper "Deep Multimodal Fusion by Channel Exchanging" (In PyTorch)
Stars: ✭ 112 (-13.18%)
Mutual labels: semantic-segmentation
Iros20 6d Pose Tracking
[IROS 2020] se(3)-TrackNet: Data-driven 6D Pose Tracking by Calibrating Image Residuals in Synthetic Domains
Stars: ✭ 113 (-12.4%)
Mutual labels: domain-adaptation
Semantic Segmentation Of Remote Sensing Images
遥感图像的语义分割，基于深度学习，在Tensorflow框架下，利用TF.Keras，运行环境TF2.0+
Stars: ✭ 125 (-3.1%)
Mutual labels: semantic-segmentation
Generate to adapt
Implementation of "Generate To Adapt: Aligning Domains using Generative Adversarial Networks"
Stars: ✭ 120 (-6.98%)
Mutual labels: domain-adaptation
Autoannotationtool
A label tool aim to reduce semantic segmentation label time, rectangle and polygon annotation is supported
Stars: ✭ 113 (-12.4%)
Mutual labels: semantic-segmentation
Dabnet
Depth-wise Asymmetric Bottleneck for Real-time Semantic Segmentation (BMVC2019)
Stars: ✭ 109 (-15.5%)
Mutual labels: semantic-segmentation
Convolutional Handwriting Gan
ScrabbleGAN: Semi-Supervised Varying Length Handwritten Text Generation (CVPR20)
Stars: ✭ 107 (-17.05%)
Mutual labels: domain-adaptation
DISE-Domain-Invariant-Structure-Extraction
Pytorch Implementation of the paper All about Structure: Adapting Structural Information across Domains for Boosting Semantic Segmentation, CVPR 2019.
| Introduction video | Paper (ArXiv) | Project Page |
|---|
Paper
All about Structure: Adapting Structural Information across Domains for Boosting Semantic Segmentation
Wei-Lun Chang*, Hui-Po Wang*, Wen-Hsiao Peng, Wei-Chen Chiu (*contribute equally)
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019.
@inproceedings{chang2019all,
title={All about Structure: Adapting Structural Information across Domains for Boosting Semantic Segmentation},
author={Chang, Wei-Lun and Wang, Hui-Po and Peng, Wen-Hsiao and Chiu, Wei-Chen},
booktitle={IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
year={2019}
}
Example Results
Quantitative Results
Prerequisite
- Pytorch 0.3.1
- Nvidia GPU with at least 16 GB memory
Installation
git clone https://github.com/a514514772/DISE-Domain-Invariant-Structure-Extraction.git
Datasets
- Download the GTA5 Dataset as the source domain and unzip it to
/data - Download the Cityscapes Dataset as the target domain and unzip it to
/data
The structure of /data may look like this:
├── data
├── Cityscapes
│ ├── gtFine
│ └── leftImg8bit
├── GTA5
├── images
└── labels
Usage
Pretrained Weights
Example Training Script: GTA5 to Cityscapes
python train_dise_gta2city.py --gta5_data_path /data/GTA5 --city_data_path /data/Cityscapes
Example Testing Script:
Note that, to test performance on the testing set, we provide scripts to generate 1024x2048 outputs which are compatible with the testing server.
python evaluate.py ./weights --city_data_path /data/Cityscapes
More options
python train_dise_gta2city.py -h
usage: train_dise_gta2city.py [-h] [--dump_logs DUMP_LOGS] [--log_dir LOG_DIR] [--gen_img_dir GEN_IMG_DIR]
[--gta5_data_path GTA5_DATA_PATH] [--city_data_path CITY_DATA_PATH]
[--data_list_path_gta5 DATA_LIST_PATH_GTA5]
[--data_list_path_city_img DATA_LIST_PATH_CITY_IMG]
[--data_list_path_city_lbl DATA_LIST_PATH_CITY_LBL]
[--data_list_path_val_img DATA_LIST_PATH_VAL_IMG]
[--data_list_path_val_lbl DATA_LIST_PATH_VAL_LBL]
[--cuda_device_id CUDA_DEVICE_ID [CUDA_DEVICE_ID ...]]
Domain Invariant Structure Extraction (DISE) for unsupervised domain adaptation for semantic segmentation
Acknowledgement
We implement this project heavily based on AdaptSeg proposed by Tsai et el..
Note that the project description data, including the texts, logos, images, and/or trademarks,
for each open source project belongs to its rightful owner.
If you wish to add or remove any projects, please contact us at [email protected].