English
 
Help Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT
 
 
DownloadE-Mail
 Local TagsRelease HistoryDetailsSummary
  DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation

Hoyer, L., Dai, D., & Van Gool, L. (2022). DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation. In IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 9914-9925). Piscataway, NJ: IEEE. doi:10.1109/CVPR52688.2022.00969.

Item is

 

show all hide all

Basic

show hide
Item Permalink: https://hdl.handle.net/21.11116/0000-000A-16B5-1 Version Permalink: https://hdl.handle.net/21.11116/0000-000C-2B17-B
Genre: Conference Paper
Latex : {DAFormer}: {I}mproving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation

Files

show Files
hide Files
:
arXiv:2111.14887.pdf (Preprint), 9MB
 
File Permalink:
-
Name:
arXiv:2111.14887.pdf
Description:
File downloaded from arXiv at 2022-03-09 14:42
OA-Status:
Visibility:
Private
MIME-Type / Checksum:
application/pdf
Technical Metadata:
Copyright Date:
-
Copyright Info:
-
License:
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
:
Hoyer_DAFormer_Improving_Network_Architectures_and_Training_Strategies_for_Domain-Adaptive_Semantic_CVPR_2022_paper.pdf (Preprint), 737KB
View   Save
File Permalink:
https://hdl.handle.net/21.11116/0000-000C-1395-6
Name:
Hoyer_DAFormer_Improving_Network_Architectures_and_Training_Strategies_for_Domain-Adaptive_Semantic_CVPR_2022_paper.pdf
Description:
-
OA-Status:
Green
Visibility:
Public
MIME-Type / Checksum:
application/pdf / [MD5]
Technical Metadata:
View
Copyright Date:
-
Copyright Info:
These CVPR 2022 papers are the Open Access versions, provided by the Computer Vision Foundation. Except for the watermark, they are identical to the accepted versions; the final published version of the proceedings is available on IEEE Xplore. This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. © 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
License:
-

Locators

show

Creators

show
hide
 Creators:
Hoyer, Lukas1, Author
Dai, Dengxin2, Author           
Van Gool, Luc1, Author
Affiliations:
1External Organizations, ou_persistent22              
2Computer Vision and Machine Learning, MPI for Informatics, Max Planck Society, ou_1116547              

Content

show
hide
Free keywords: Computer Science, Computer Vision and Pattern Recognition, cs.CV
 Abstract: As acquiring pixel-wise annotations of real-world images for semantic
segmentation is a costly process, a model can instead be trained with more
accessible synthetic data and adapted to real images without requiring their
annotations. This process is studied in unsupervised domain adaptation (UDA).
Even though a large number of methods propose new adaptation strategies, they
are mostly based on outdated network architectures. As the influence of recent
network architectures has not been systematically studied, we first benchmark
different network architectures for UDA and then propose a novel UDA method,
DAFormer, based on the benchmark results. The DAFormer network consists of a
Transformer encoder and a multi-level context-aware feature fusion decoder. It
is enabled by three simple but crucial training strategies to stabilize the
training and to avoid overfitting DAFormer to the source domain: While the Rare
Class Sampling on the source domain improves the quality of pseudo-labels by
mitigating the confirmation bias of self-training towards common classes, the
Thing-Class ImageNet Feature Distance and a learning rate warmup promote
feature transfer from ImageNet pretraining. DAFormer significantly improves the
state-of-the-art performance by 10.8 mIoU for GTA->Cityscapes and 5.4 mIoU for
Synthia->Cityscapes and enables learning even difficult classes such as train,
bus, and truck well. The implementation is available at
https://github.com/lhoyer/DAFormer.

Details

show
hide
Language(s): eng - English
 Dates: 2021-11-2920222022
 Publication Status: Published online
 Pages: -
 Publishing info: -
 Table of Contents: -
 Rev. Type: -
 Identifiers: BibTex Citekey: Hoyer_CVPR2022
DOI: 10.1109/CVPR52688.2022.00969
 Degree: -

Event

show
hide
Title: 35th IEEE/CVF Conference on Computer Vision and Pattern Recognition
Place of Event: New Orleans, LA, USA
Start-/End Date: 2022-06-19 - 2022-06-24

Legal Case

show

Project information

show

Source 1

show
hide
Title: IEEE/CVF Conference on Computer Vision and Pattern Recognition
  Abbreviation : CVPR 2022
Source Genre: Proceedings
 Creator(s):
Affiliations:
Publ. Info: Piscataway, NJ : IEEE
Pages: - Volume / Issue: - Sequence Number: - Start / End Page: 9914 - 9925 Identifier: ISBN: 978-1-6654-6946-3