Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Domain Adaptation for Semantic Segmentation with Maximum Squares Loss

About

Deep neural networks for semantic segmentation always require a large number of samples with pixel-level labels, which becomes the major difficulty in their real-world applications. To reduce the labeling cost, unsupervised domain adaptation (UDA) approaches are proposed to transfer knowledge from labeled synthesized datasets to unlabeled real-world datasets. Recently, some semi-supervised learning methods have been applied to UDA and achieved state-of-the-art performance. One of the most popular approaches in semi-supervised learning is the entropy minimization method. However, when applying the entropy minimization to UDA for semantic segmentation, the gradient of the entropy is biased towards samples that are easy to transfer. To balance the gradient of well-classified target samples, we propose the maximum squares loss. Our maximum squares loss prevents the training process being dominated by easy-to-transfer samples in the target domain. Besides, we introduce the image-wise weighting ratio to alleviate the class imbalance in the unlabeled target domain. Both synthetic-to-real and cross-city adaptation experiments demonstrate the effectiveness of our proposed approach. The code is released at https://github. com/ZJULearning/MaxSquareLoss.

Minghao Chen, Hongyang Xue, Deng Cai• 2019

Related benchmarks

TaskDatasetResultRank
Semantic segmentationCityscapes (val)
mIoU46.4
572
Semantic segmentationGTA5 → Cityscapes (val)
mIoU46.4
533
Semantic segmentationSYNTHIA to Cityscapes (val)
Rider IoU20.5
435
Semantic segmentationCityscapes GTA5 to Cityscapes adaptation (val)
mIoU (Overall)46.4
352
Semantic segmentationGTA5 to Cityscapes (test)
mIoU46.4
151
Semantic segmentationSYNTHIA to Cityscapes
Road IoU82.9
150
Semantic segmentationSynthia to Cityscapes (test)
Road IoU82.9
138
Semantic segmentationCityscapes adaptation from Synthia 1.0 (val)
Person IoU58.5
114
Semantic segmentationGTA to Cityscapes
Road IoU88.1
72
Semantic segmentationSYNTHIA-to-Cityscapes (SYN2CS) 16 classes (val)
IoU41.4
50
Showing 10 of 16 rows

Other info

Follow for update