# Exploiting Inter-pixel Correlations in Unsupervised Domain Adaptation for Semantic Segmentation

Source: [Exploiting Inter-pixel Correlations in Unsupervised Domain Adaptation for Semantic Segmentation](<https://www.sota2.com/research/paper/68fa6b0e1e209131cea2bcd4>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-27T06:10:35.614Z

Authors: Inseop Chung, Jayeon Yoo, Nojun Kwak

Published: 2021-10-21T06:11:44.000Z

## Abstract

"Self-training" has become a dominant method for semantic segmentation via unsupervised domain adaptation (UDA). It creates a set of pseudo labels for the target domain to give explicit supervision. However, the pseudo labels are noisy, sparse and do not provide any information about inter-pixel correlations. We regard inter-pixel correlation quite important because semantic segmentation is a task of predicting highly structured pixel-level outputs. Therefore, in this paper, we propose a method of transferring the inter-pixel correlations from the source domain to the target domain via a self-attention module. The module takes the prediction of the segmentation network as an input and creates a self-attended prediction that correlates similar pixels. The module is trained only on the source domain to learn the domain-invariant inter-pixel correlations, then later, it is used to train the segmentation network on the target domain. The network learns not only from the pseudo labels but also by following the output of the self-attention module which provides additional knowledge about the inter-pixel correlations. Through extensive experiments, we show that our method significantly improves the performance on two standard UDA benchmarks and also can be combined with recent state-of-the-art method to achieve better performance.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Semantic Segmentation on SYNTHIA to Cityscapes (val)](<https://www.sota2.com/research/sota/semantic-segmentation-on-synthia-to-cityscapes-val.md>): Semantic segmentation; SYNTHIA to Cityscapes (val); Rider IoU: 31.8.

- [Semantic Segmentation on GTA5 → Cityscapes (val)](<https://www.sota2.com/research/sota/semantic-segmentation-on-gta5-cityscapes-val.md>): Semantic segmentation; GTA5 → Cityscapes (val); mIoU: 52.2.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2110.10916>)

[PDF](<https://arxiv.org/pdf/2110.10916.pdf>)