Domain Adaptive Semantic Segmentation with Self-Supervised Depth Estimation Article Swipe

MAKE IMPACT

Qin Wang , Dengxin Dai , Lukas Hoyer , Olga Fink , Luc Van Gool · YOU? · · 2021 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2104.13613

Domain adaptation for semantic segmentation aims to improve the model performance in the presence of a distribution shift between source and target domain. Leveraging the supervision from auxiliary tasks~(such as depth estimation) has the potential to heal this shift because many visual tasks are closely related to each other. However, such a supervision is not always available. In this work, we leverage the guidance from self-supervised depth estimation, which is available on both domains, to bridge the domain gap. On the one hand, we propose to explicitly learn the task feature correlation to strengthen the target semantic predictions with the help of target depth estimation. On the other hand, we use the depth prediction discrepancy from source and target depth decoders to approximate the pixel-wise adaptation difficulty. The adaptation difficulty, inferred from depth, is then used to refine the target semantic segmentation pseudo-labels. The proposed method can be easily implemented into existing segmentation frameworks. We demonstrate the effectiveness of our approach on the benchmark tasks SYNTHIA-to-Cityscapes and GTA-to-Cityscapes, on which we achieve the new state-of-the-art performance of $55.0\%$ and $56.6\%$, respectively. Our code is available at \url{https://qin.ee/corda}.

Type

preprint

Language

en

Landing Page

http://arxiv.org/abs/2104.13613

PDF

https://arxiv.org/pdf/2104.13613

OA Status

green

Cited By

6

References

58

Related Works

10

OpenAlex ID

https://openalex.org/W3158314417

Raw OpenAlex JSON

OpenAlex ID

https://openalex.org/W3158314417

Canonical identifier for this work in OpenAlex

DOI

https://doi.org/10.48550/arxiv.2104.13613

Digital Object Identifier

Title

Domain Adaptive Semantic Segmentation with Self-Supervised Depth Estimation

Work title

Type

preprint

OpenAlex work type

Language

en

Primary language

Publication year

2021

Year of publication

Publication date

2021-04-28

Full publication date if available

Authors

Qin Wang, Dengxin Dai, Lukas Hoyer, Olga Fink, Luc Van Gool

List of authors in order

Landing page

https://arxiv.org/abs/2104.13613

Publisher landing page

PDF URL

https://arxiv.org/pdf/2104.13613

Direct link to full text PDF

Open access

Yes

Whether a free full text is available

OA status

green

Open access status per OpenAlex

OA URL

https://arxiv.org/pdf/2104.13613

Direct OA link when available

Concepts

Segmentation, Computer science, Leverage (statistics), Artificial intelligence, Benchmark (surveying), Domain adaptation, Code (set theory), Adaptation (eye), Domain (mathematical analysis), Feature (linguistics), Task (project management), Bridge (graph theory), Pattern recognition (psychology), Machine learning, Mathematics, Medicine, Physics, Economics, Set (abstract data type), Management, Programming language, Linguistics, Geodesy, Optics, Philosophy, Mathematical analysis, Classifier (UML), Internal medicine, Geography

Top concepts (fields/topics) attached by OpenAlex

Cited by

6

Total citation count in OpenAlex

Citations by year (recent)

2023: 6

Per-year citation counts (last 5 years)

References (count)

58

Number of works referenced by this work

Related works (count)

10

Other works algorithmically related by OpenAlex

Full payload