Awesome

Paper

Generalization by Adaptation: Diffusion-Based Domain Extension for Domain-Generalized Semantic Segmentation<br> Joshua Niemeijer*, Manuel Schwonberg*, Jan-Aike Termöhlen*, Nico M. Schmidt, and Tim Fingscheidt<br> Winter Conference on Applications of Computer Vision (WACV) 2024<br> (* indicates equal contribution)

The full code will be published soon.

Installation

To utilize DIDEX please follow the following steps:

For the creation of the pseudo target domain we build on the following repos:

1. https://github.com/Stability-AI/stablediffusion.git
2. https://github.com/lllyasviel/ControlNet.git

For the adaptation to the pseudo target domain we utilize the following repo:

1. https://github.com/lhoyer/MIC.git

To utilize our code please set up the repos following the descriptions they provide.

Diffusion-Based Domain Extension (Pseudo-Target Domain Generation)

To create the Pseudo target domains please utilize the scripts in the folder dataset_creation.

Adaptation To Pseudo-Target Domain

To train the model for domain generalization please utilize the scripts in generalization_experiments

Datasets

We used the dataset structure ...

Evaluation

BibTeX

@article{Niemeijer2023DIDEX,,
  author        = {Niemeijer, Joshua and Schwonberg, Manuel and Termöhlen, Jan-Aike and Schmidt, Nico M. and Fingscheidt, Tim},
  title         = {{Generalization by Adaptation: Diffusion-Based Domain Extension for Domain-Generalized Semantic Segmentation}},
  year          = {2023},
  month         = dec,
  pages         = {1--16},
  eprint        = {2312.01850},
  archivePrefix = {arXiv}
}