IAPC

Towards Source-free Domain Adaptive Semantic Segmentation via Importance-aware and Prototype-contrast Learning

Paper

Code for our paper "Towards Source-free Domain Adaptive Semantic Segmentation via Importance-aware and Prototype-contrast Learning" (Accepted by IEEE Transactions on Intelligent Vehicles)

In this paper, we propose an end-to-end source-free domain adaptation semantic segmentation method via Importance-Aware and Prototype-Contrast (IAPC) learning. The proposed IAPC framework effectively extracts domain-invariant knowledge from the well-trained source model and learns domain-specific knowledge from the unlabeled target domain.

Code

List

• Code
• Checkpoints and prediction maps

Prerequisites

The code is implemented with Python(3.6) and Pytorch(1.7).

Datasets

• Download Cityscapes datasets

Training

python generate_plabel_cityscapes.py
python train.py

Testing

python evaluate_cityscapes.py

Results

	Model	mIoU	mIoU*
GTA5-to-Cityscapes	Source Only	36.6	-
	IAPC	49.4	-
Synthia-to-Cityscapes	Source Only	35.2	40.5
	IAPC	45.3	52.6

IAPC for Object Detection

please refer to the subdirectory OD