Phase Consistent Ecological Domain Adaptation

By Yanchao Yang, Dong Lao, Ganesh Sundaramoorthi, Stefano Soatto. 2020

130 citations recorded by the source corpus.

We introduce two criteria to regularize the optimization involved in learning a classifier in a domain where no annotated data are available, leveraging annotated data in a different domain, a problem known as unsupervised domain adaptation. We focus on the task of semantic segmentation, where annotated synthetic data are aplenty, but annotating real data is laborious. The first criterion, inspired by visual psychophysics, is that the map between the two image domains be phase-preserving. This restricts the set of possible learned maps, while enabling enough flexibility to transfer semantic information. The second criterion aims to leverage ecological statistics, or regularities in the scene which are manifest in any image of it, regardless of the characteristics of the illuminant or the imaging sensor. It is implemented using a deep neural network that scores the likelihood of each possible segmentation given a single un-annotated image. Incorporating these two priors in a standard domain adaptation framework improves performance across the board in the most common unsupervised domain adaptation benchmarks for semantic segmentation.

Subfields: Representation & Self-Supervised Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Representation & Self-Supervised Learning

1. Dacheng Tao
2. Mingming Gong
3. Kun Zhang
4. Mohammad Rostami
5. Serban Stan
6. Yanchao Yang
7. Peilin Zhao
8. Jing Qin

Citations in the corpus

• Unsupervised Model Adaptation for Continual Semantic Segmentation (2020, background): “2018), TGCF-DA+SE (Choi, Kim, and Kim 2019), PCEDA (Yang et al. 2020), and CyCADA (Hoffman et al.”
• FDA: Fourier Domain Adaptation for Semantic Segmentation (2020, background): “Note in the concurrent work [48], a transformer network trained with phase preservation as a constraint also generates domain aligned images that maintain semantic content in the source images.”
• FedDG: Federated Domain Generalization on Medical Image Segmentation via Episodic Learning in Continuous Frequency Space (2021, background): “, style) and highlevel semantics of an image can be respectively captured by amplitude and phase spectrum in the frequency space, as revealed by visual psychophysics [13, 42, 57].”
• A Fourier-based Framework for Domain Generalization (2021, unknown): “[47] proposes to keep a phase consistency in source-target image translations, which is shown to be a better choice than the commonly used cycle consistency [15] for segmentation tasks under DA scenario.”
• Exploring Robustness of Unsupervised Domain Adaptation in Semantic Segmentation (2021, background): “, FDA [48] and PCEDA [47] suffer from mIoU drop of 31.”
• Alleviating Semantics Distortion in Unsupervised Low-Level Image-to-Image Translation via Structure Consistency Constraint (2022, methodology): “Moreover, [5, 40, 48, 56, 57, 60, 68, 71] use the attention-based/ pretrained model or pre-define functions to preserve the semantics during translation.”