Condition-Invariant Semantic Segmentation

05/27/2023
by   Christos Sakaridis, et al.
0

Adaptation of semantic segmentation networks to different visual conditions from those for which ground-truth annotations are available at training is vital for robust perception in autonomous cars and robots. However, previous work has shown that most feature-level adaptation methods, which employ adversarial training and are validated on synthetic-to-real adaptation, provide marginal gains in normal-to-adverse condition-level adaptation, being outperformed by simple pixel-level adaptation via stylization. Motivated by these findings, we propose to leverage stylization in performing feature-level adaptation by aligning the deep features extracted by the encoder of the network from the original and the stylized view of each input image with a novel feature invariance loss. In this way, we encourage the encoder to extract features that are invariant to the style of the input, allowing the decoder to focus on parsing these features and not on further abstracting from the specific style of the input. We implement our method, named Condition-Invariant Semantic Segmentation (CISS), on the top-performing domain adaptation architecture and demonstrate a significant improvement over previous state-of-the-art methods both on Cityscapes→ACDC and Cityscapes→Dark Zurich adaptation. In particular, CISS is ranked first among all published unsupervised domain adaptation methods on the public ACDC leaderboard. Our method is also shown to generalize well to domains unseen during training, outperforming competing domain adaptation approaches on BDD100K-night and Nighttime Driving. Code is publicly available at https://github.com/SysCV/CISS .

READ FULL TEXT

page 1

page 3

page 7

research
07/14/2022

Refign: Align and Refine for Adaptation of Semantic Segmentation to Adverse Conditions

Due to the scarcity of dense pixel-level semantic annotations for images...
research
04/27/2023

EDAPS: Enhanced Domain-Adaptive Panoptic Segmentation

With autonomous industries on the rise, domain adaptation of the visual ...
research
03/07/2022

Semantic Segmentation in Art Paintings

Semantic segmentation is a difficult task even when trained in a supervi...
research
04/30/2021

Reproducibility of "FDA: Fourier Domain Adaptation forSemantic Segmentation

The following paper is a reproducibility report for "FDA: Fourier Domain...
research
05/13/2019

Weakly-supervised Caricature Face Parsing through Domain Adaptation

A caricature is an artistic form of a person's picture in which certain ...
research
03/09/2023

Contrastive Model Adaptation for Cross-Condition Robustness in Semantic Segmentation

Standard unsupervised domain adaptation methods adapt models from a sour...
research
11/29/2021

DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation

As acquiring pixel-wise annotations of real-world images for semantic se...

Please sign up or login with your details

Forgot password? Click here to reset