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: Nov 14, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

Category-Level Adversarial Adaptation for Semantic Segmentation Using Purified Features.

Yawei Luo, Ping Liu, Liang Zheng

IEEE Transactions on Pattern Analysis and Machine Intelligence

|March 8, 2021

Summary

This summary is machine-generated.

Related Concept Videos

Survival Tree

Survival Tree

Survival trees are a non-parametric method used in survival analysis to model the relationship between a set of covariates and the time until an event of interest occurs, often referred to as the "time-to-event" or "survival time." This method is particularly useful when dealing with censored data, where the event has not occurred for some individuals by the end of the study period, or when the exact time of the event is unknown.
Building a Survival Tree
Constructing a...

You might also read

Related Articles

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

Sort by

Same author

Multimodal anti-Helicobacter pylori effects of Lactobacillus casei HY001: Evidence from in vitro and in vivo studies.

PloS one·2026

Same author

Trichosanthis peel injection improves functional recovery after spinal cord injury associated with the modulation of neuroinflammation and oxidative stress.

The journal of spinal cord medicine·2026

Same author

Rat models of thrombotic thrombocytopenic purpura reveal crucial role of placental ADAMTS13 in perinatal survival.

Blood advances·2026

Same author

Generating adamts13<sup>-/-</sup> Zebrafish via CRISPR/Cas9 Gene Editing.

Methods in molecular biology (Clifton, N.J.)·2026

Same author

PRIME-HFrEF Trial: a randomized, double-blind, multi-dose umbilical cord-derived mesenchymal stem cell regimen for heart failure.

Signal transduction and targeted therapy·2026

Same author

Visual perception preferences in Lingnan gardens: a semantic differential survey with a convenience sample.

Frontiers in psychology·2026

Same journal

TraGraph-GS: Trajectory Graph-based Gaussian Splatting for Arbitrary Large-Scale Scene Rendering.

IEEE transactions on pattern analysis and machine intelligence·2026

Same journal

SWIFT: A Small-World Interaction Framework for Flow-Aware Trajectory Prediction in Autonomous Driving.

IEEE transactions on pattern analysis and machine intelligence·2026

Same journal

HardFlow: Hard-Constrained Sampling for Flow-Matching Models Via Trajectory Optimization.

IEEE transactions on pattern analysis and machine intelligence·2026

Same journal

Industrial Brain: Self-Evolving Neuro-Symbolic Autonomy with Causal Resilience for Cyber-Physical Systems.

IEEE transactions on pattern analysis and machine intelligence·2026

Same journal

Adaptive Hardness-Driven Dictionary Distillation for Incomplete Streaming View Clustering.

IEEE transactions on pattern analysis and machine intelligence·2026

Same journal

Mixture of Global and Local Experts with Diffusion Transformer for Controllable Face Generation.

IEEE transactions on pattern analysis and machine intelligence·2026

See all related articles

This study introduces a new method for unsupervised domain adaptive semantic segmentation, improving generalization across different datasets. The Category-level Alignment Network (CLAN) effectively reduces domain shift by aligning both global and local distributions.

Area of Science:

Computer Vision
Machine Learning
Artificial Intelligence

Background:

Unsupervised domain adaptive semantic segmentation aims to generalize models to new domains without labeled data.
Current methods often align global feature distributions but neglect local joint distributions, limiting performance.
Noisy, irrelevant features can hinder effective domain alignment.

Purpose of the Study:

To develop a novel approach for unsupervised domain adaptive semantic segmentation that addresses limitations of existing methods.
To improve the generalization capability of semantic segmentation models across diverse domains.
To disentangle task-relevant features from noisy factors for more robust domain adaptation.

Main Methods:

Introduced Significance-aware Information Bottleneck (SIB) to filter out noisy features.

Related Experiment Videos

Last Updated: Nov 14, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

Developed Category-level Alignment (CLA) for simultaneous global and local distribution alignment.

Proposed a purified embedding-based category-level adversarial network (CLAN).

Main Results:

CLAN effectively disentangles noisy factors, suppressing their influence on the target task.
The method achieves simultaneous alignment of global marginal and local joint distributions.
State-of-the-art segmentation accuracy was validated across three challenging domain adaptation tasks (GTA5 → Cityscapes, SYNTHIA → Cityscapes, Cross Season).

Conclusions:

The proposed CLAN method offers a significant advancement in unsupervised domain adaptive semantic segmentation.
Simultaneous alignment of global and local distributions, coupled with noise suppression, is crucial for optimal domain adaptation.
The approach demonstrates strong generalization capabilities, achieving competitive results on benchmark datasets.