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

Classes Matter: A Fine-grained Adversarial Approach to Cross-domain Semantic Segmentation

About

Despite great progress in supervised semantic segmentation,a large performance drop is usually observed when deploying the model in the wild. Domain adaptation methods tackle the issue by aligning the source domain and the target domain. However, most existing methods attempt to perform the alignment from a holistic view, ignoring the underlying class-level data structure in the target domain. To fully exploit the supervision in the source domain, we propose a fine-grained adversarial learning strategy for class-level feature alignment while preserving the internal structure of semantics across domains. We adopt a fine-grained domain discriminator that not only plays as a domain distinguisher, but also differentiates domains at class level. The traditional binary domain labels are also generalized to domain encodings as the supervision signal to guide the fine-grained feature alignment. An analysis with Class Center Distance (CCD) validates that our fine-grained adversarial strategy achieves better class-level alignment compared to other state-of-the-art methods. Our method is easy to implement and its effectiveness is evaluated on three classical domain adaptation tasks, i.e., GTA5 to Cityscapes, SYNTHIA to Cityscapes and Cityscapes to Cross-City. Large performance gains show that our method outperforms other global feature alignment based and class-wise alignment based counterparts. The code is publicly available at https://github.com/JDAI-CV/FADA.

Haoran Wang, Tong Shen, Wei Zhang, Lingyu Duan, Tao Mei• 2020

Related benchmarks

TaskDatasetResultRank
Semantic segmentationCityscapes (test)
mIoU52.4
1145
Semantic segmentationGTA5 → Cityscapes (val)
mIoU50.6
533
Semantic segmentationSYNTHIA to Cityscapes (val)
Rider IoU85.4
435
Semantic segmentationCityscapes GTA5 to Cityscapes adaptation (val)
mIoU (Overall)49.2
352
Semantic segmentationSYNTHIA to Cityscapes
Road IoU84.5
150
Semantic segmentationSynthia to Cityscapes (test)
Road IoU57.3
138
Semantic segmentationGTA5 to Cityscapes 1.0 (val)
Road IoU92.5
98
Semantic segmentationGTA to Cityscapes
Road IoU91
72
Semantic segmentationSYNTHIA-to-Cityscapes (SYN2CS) 16 classes (val)--
50
Semantic segmentationCityscapes GTA5 source 1.0 (val)
mIoU50.1
49
Showing 10 of 22 rows

Other info

Code

Follow for update