Skip to content
/ MDIL-SS Public

Code repository for the work "Multi-Domain Incremental Learning for Semantic Segmentation", accepted at WACV 2022

52 stars 5 forks Branches Tags Activity
Star
Notifications You must be signed in to change notification settings

prachigarg23/MDIL-SS

BranchesTags

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

16 Commits
models
models
 
 
trained_models
trained_models
 
 
trainer_files
trainer_files
 
 
Evaluation_Notebook.ipynb
Evaluation_Notebook.ipynb
 
 
Plot_Tsne_Notebook.ipynb
Plot_Tsne_Notebook.ipynb
 
 
README.md
README.md
 
 
cal_class_weights.py
cal_class_weights.py
 
 
dataset.py
dataset.py
 
 
dataset_custom.py
dataset_custom.py
 
 
final-main-diagram-wacv1.png
final-main-diagram-wacv1.png
 
 
iouEval.py
iouEval.py
 
 
main_FT2_flexible_new.py
main_FT2_flexible_new.py
 
 
main_ftp1_enc_newbn.py
main_ftp1_enc_newbn.py
 
 
train_RAPFT_step1.py
train_RAPFT_step1.py
 
 
train_multi_task.py
train_multi_task.py
 
 
train_new_task_step2.py
train_new_task_step2.py
 
 
train_new_task_step3.py
train_new_task_step3.py
 
 
transform.py
transform.py
 
 

Repository files navigation

Multi-Domain Incremental Learning for Semantic Segmentation

This is the Pytorch implementation of our work "Multi-Domain Incremental Learning for Semantic Segmentation", accepted at WACV 2022 (Algorithms Track)

Full paper: https://arxiv.org/abs/2110.12205

[New] Model checkpoints and evaluation notebook now out for easy reproducibility!

image

Requirements

  • Python 3.6
  • Pytorch: Make sure to install the Pytorch version for Python 3.6 with CUDA support (This code has been run with CUDA 10.2)
  • Models can take upto 40 hours on 2 Nvidia GeForce GTX 1080 Ti GPUs for Step 2; and upto 90 hours on 4 Nvidia GeForce GTX 1080 Ti GPUs for Step 3.

Datasets

Preprocessing IDD: convert polygon labels to segmentation masks:

  1. Clone