2021
DOI: 10.1109/tip.2021.3096334
|View full text |Cite
|
Sign up to set email alerts
|

Global and Local Texture Randomization for Synthetic-to-Real Semantic Segmentation

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
2

Citation Types

0
26
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
3
2
2

Relationship

1
6

Authors

Journals

citations
Cited by 59 publications
(26 citation statements)
references
References 53 publications
0
26
0
Order By: Relevance
“…Domain generalization has been mostly explored on the image classification task, and a number of approaches have been proposed using as meta-learning [2,29,30,34], adversarial training [31,33,52], autoencoders [18,33], metric learning [12,42] and data augmentation [19,74]. The research on domain generalization for semantic segmentation (DGSS) is still in its infancy, with only a few existing approaches [9,47,48,51,66]. These existing DGSS methods mainly focus on two aspects: (1) Domain Randomization and (2) Normalization and Whitening.…”
Section: Domain Generalization (Dg)mentioning
confidence: 99%
See 2 more Smart Citations
“…Domain generalization has been mostly explored on the image classification task, and a number of approaches have been proposed using as meta-learning [2,29,30,34], adversarial training [31,33,52], autoencoders [18,33], metric learning [12,42] and data augmentation [19,74]. The research on domain generalization for semantic segmentation (DGSS) is still in its infancy, with only a few existing approaches [9,47,48,51,66]. These existing DGSS methods mainly focus on two aspects: (1) Domain Randomization and (2) Normalization and Whitening.…”
Section: Domain Generalization (Dg)mentioning
confidence: 99%
“…[66] leverages the advanced image-to-image translation to transfer a source domain image to multiple styles aiming to learn a model with high generalizability. Similarly, GTR [51] randomizes the synthetic images with the styles of unreal paintings in order to learn domain-invariant representations. Normalization and Whitening Methods apply different normalization techniques such as Instance Normalization (IN) [61] or whitening [48].…”
Section: Domain Generalization (Dg)mentioning
confidence: 99%
See 1 more Smart Citation
“…Unlike DA, DG does not have access to the target domains during the learning process. To make reliable predictions on various 'unknown' target domains, most existing studies focus on whitening [7], normalizing [41], and diversifying [22,45,62] styles to avoid overfitting to the style of the source domain. This paper focuses on extending both the content and style of the source domain to the wild [11], enabling networks to learn domain-generalized semantic features from diversified contents and styles.…”
Section: Domain Adaptation and Generalizationmentioning
confidence: 99%
“…In this paper, we propose a new domain generalized semantic segmentation network called WildNet, which learns the domain-generalized semantic feature by 'extending' both content and style to the wild. Although some previous works [22,45,62] utilized various styles from the wild, e.g., ImageNet [11] for real styles and Painter by Numbers [38] for unreal styles, they overlooked that the high generalization ability comes from learning not only various styles but also various contents. In contrast to previous studies, our main idea is to naturally learn domain-generalized semantic information by leveraging a variety of contents and styles from the wild, without forcing whitening on domainspecific styles.…”
Section: Introductionmentioning
confidence: 99%