Some useful Domain adaptation code
This repository collect some useful code fun others. And we will implement them in the future. So you can use them easily.
Learning to Adapt Structured Output Space for Semantic Segmentation
- paper link: https://arxiv.org/abs/1802.10349
- code: provided in Folder: Adapt Structured Output
- main code forked from https://github.com/wasidennis/AdaptSegNet
- tutorial Video
No More Discrimination: Cross City Adaptation of Road Scene Segmenters
- paper link: https://arxiv.org/abs/1704.08509
- code: provided in Folder: Adapt_Road_Scene
- tutorial Video
FCNs in the Wild: Pixel-level Adversarial and Constraint-based Adaptation
- paper link: https://arxiv.org/abs/1612.02649
- code: provided in Folder: FCNs_Wild
- tutorial Video
Maximum Classifier Discrepancy for Domain Adaptation with Semantic Segmentation
- paper link: https://arxiv.org/abs/1712.02560
- code: provided in Folder: MCD_DA_seg
- main code forked from https://github.com/mil-tokyo/MCD_DA
- tutorial Video
- Learning to Adapt Structured Output Space for Semantic Segmentation, Wei-Chih Hung.
- No More Discrimination: Cross City Adaptation of Road Scene Segmenters, Yu-Ting Chen.
- Maximum Classifier Discrepancy for Domain Adaptation with Semantic Segmentation, Kuniaki Saito.
-
Synthia Dataset ( Download the subset SYNTHIA-RAND-CITYSCAPES)
-
- contains four subsets --- Taipei, Tokyo, Roma, Rio --- used as target domain (only testing data has annotations)