Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Jan 8, 2026

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

A3-TTA: Adaptive Anchor Alignment Test-Time Adaptation for Image Segmentation.

Jianghao Wu, Xiangde Luo, Yubo Zhou

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society

|December 22, 2025

Summary

This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Acute changes in ankle dorsiflexor strength and fNIRS-Derived cortical activation following a single session of neuromuscular electrical stimulation in healthy older adults.

Frontiers in aging·2026

Same author

Machine learning for predicting surgical difficulty of laparoscopic total mesorectal excision for rectal cancer: integrating MR-based pelvimetry and peritoneal reflection.

Frontiers in medicine·2026

Same author

SUDA: Simultaneous unsupervised knowledge distillation and adaptation of foundation models for efficient pathological image analysis.

Medical image analysis·2026

Same author

Integrated multi-omics analysis reveals the molecular mechanism underlying poplar 107 rootstock-mediated regulation of <i>Populus tomentosa</i> scion growth.

Horticulture research·2026

Same author

Cellular architecture and neighborhood-informed virtual spatial tumor profiling from histopathology.

Cell·2026

Same author

Myocardial salvage and functional recovery prediction: T1 rho vs conventional CMR in acute myocardial infarction.

Magnetic resonance imaging·2026

Same journal

Mask-guided Asymmetric Contrastive and Semantic Alignment for Unsupervised Person Re-Identification.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026

Same journal

Hyperbolic Cycle Alignment for Infrared-Visible Image Fusion.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026

Same journal

Learning Gaze Synthesizer via 3D-eye Controlled Diffusion and Cross-domain Feature Alignment.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026

Same journal

Underlying Semantic Diffusion for Effective and Efficient In-Context Learning.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026

Same journal

DiffRES: Unleashing Text-to-Image Diffusion Models for Generative Referring Expression Segmentation without Information Leakage.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026

Same journal

Location Matters: Frequency-Spatial Dual Space Adaptation for Cross-Domain Few-Shot Segmentation.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026

See all related articles

A novel Test-Time Adaptation (TTA) framework, A3-TTA, generates reliable pseudo-labels using anchor-guided supervision. This approach significantly enhances image segmentation performance under domain shift, outperforming existing methods.

Area of Science:

Computer Vision
Machine Learning
Medical Imaging

Background:

Test-Time Adaptation (TTA) enables image segmentation model deployment under domain shift without retraining.
Pseudo-labeling is a common TTA strategy, but existing methods using perturbation heuristics yield unstable training signals and error accumulation.

Purpose of the Study:

To develop a robust TTA framework that generates reliable pseudo-labels for improved image segmentation performance.
To address the instability and error accumulation issues in current pseudo-label-based TTA methods.

Main Methods:

Proposed A3-TTA framework utilizing anchor-guided supervision for reliable pseudo-label generation.
Identified well-predicted target domain images (anchors) using class compact density metric.

More Related Videos

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

Related Experiment Videos

Last Updated: Jan 8, 2026

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

Regularized pseudo-label generation via semantic consistency and boundary-aware entropy minimization.

Introduced self-adaptive exponential moving average for label noise mitigation and stable model updates.

Main Results:

A3-TTA significantly improved average Dice scores by 10.40 to 17.68 percentage points compared to the source model on medical and natural image datasets.
Outperformed several state-of-the-art TTA methods across different segmentation architectures.
Demonstrated strong performance in continual TTA with excellent anti-forgetting capabilities.

Conclusions:

A3-TTA provides a stable and effective solution for domain-adaptive image segmentation.
Anchor-guided supervision and proposed regularization techniques enhance pseudo-label reliability.
The framework shows promise for real-world deployment of segmentation models under varying data distributions.