Texture Underfitting for Domain Adaptation

08/29/2019
by   Jan-Nico Zaech, et al.
8

Comprehensive semantic segmentation is one of the key components for robust scene understanding and a requirement to enable autonomous driving. Driven by large scale datasets, convolutional neural networks show impressive results on this task. However, a segmentation algorithm generalizing to various scenes and conditions would require an enormously diverse dataset, making the labour intensive data acquisition and labeling process prohibitively expensive. Under the assumption of structural similarities between segmentation maps, domain adaptation promises to resolve this challenge by transferring knowledge from existing, potentially simulated datasets to new environments where no supervision exists. While the performance of this approach is contingent on the concept that neural networks learn a high level understanding of scene structure, recent work suggests that neural networks are biased towards overfitting to texture instead of learning structural and shape information. Considering the ideas underlying semantic segmentation, we employ random image stylization to augment the training dataset and propose a training procedure that facilitates texture underfitting to improve the performance of domain adaptation. In experiments with supervised as well as unsupervised methods for the task of synthetic-to-real domain adaptation, we show that our approach outperforms conventional training methods.

READ FULL TEXT

page 3

page 4

research
11/30/2018

ADVENT: Adversarial Entropy Minimization for Domain Adaptation in Semantic Segmentation

Semantic segmentation is a key problem for many computer vision tasks. W...
research
03/02/2020

Learning Texture Invariant Representation for Domain Adaptation of Semantic Segmentation

Since annotating pixel-level labels for semantic segmentation is laborio...
research
12/06/2021

Unsupervised Domain Adaptation for Semantic Image Segmentation: a Comprehensive Survey

Semantic segmentation plays a fundamental role in a broad variety of com...
research
11/20/2018

Sensor Adaptation for Improved Semantic Segmentation of Overhead Imagery

Semantic segmentation is a powerful method to facilitate visual scene un...
research
04/04/2023

FREDOM: Fairness Domain Adaptation Approach to Semantic Scene Understanding

Although Domain Adaptation in Semantic Scene Segmentation has shown impr...
research
12/12/2021

Image-to-Height Domain Translation for Synthetic Aperture Sonar

Observations of seabed texture with synthetic aperture sonar are depende...
research
03/29/2023

Domain Adaptive Semantic Segmentation by Optimal Transport

Scene segmentation is widely used in the field of autonomous driving for...

Please sign up or login with your details

Forgot password? Click here to reset