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

Related Concept Videos

Adhesion

Adhesion

Adhesion occurs when one type of molecule is attracted to a different molecule. Water exhibits adhesive properties in the presence of polar surfaces, such as glass or cellulose in plants. For instance, when water is poured into a glass, the positively charged hydrogen molecules of water are more attracted to the negatively charged oxygen molecules in the silica than to the oxygen in neighboring water molecules.
Capillary action is a result of water’s adhesive tendencies. When a narrow...

You might also read

Related Articles

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

Sort by

Same author

Point-to-Point Orientation Aligns P-Tailored Fe-N<sub>5</sub> Motifs within Hierarchical Carbon Nanoreactors for Superior Oxygen Electrocatalysis in Zn-Air Batteries.

Nano letters·2026

Same author

ED50 and ED95 of Remimazolam Combined with Sufentanil for Sedation During Ultrasound-Guided High-Intensity Focused Ultrasound Ablation of Uterine Fibroids: A Prospective Dose-Finding Study.

Drug design, development and therapy·2026

Same author

Caregiving burden and depressive symptoms among rural caregivers of older adults living with disabilities in China: the role of formal and informal support.

Global health action·2026

Same author

Comparing the roles of f0, speech rate, and timbre in expressing and perceiving politeness in Mandarin speech.

Phonetica·2026

Same author

Phosphoethanolamine cytidylyltransferase 2 integrates DAG metabolism and TBK1 activation to regulate antiviral innate immunity.

Proceedings of the National Academy of Sciences of the United States of America·2026

Same author

Cryo-EM Structure of the TRPC1/5 Heteromer Enables Design of Antidepressant and Anxiolytic Drug with Reduced Side Effects.

Nature communications·2026

Search research articles

Related Experiment Video

Updated: Jul 17, 2025

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

SAN: Side Adapter Network for Open-Vocabulary Semantic Segmentation.

Mengde Xu, Zheng Zhang, Fangyun Wei

IEEE Transactions on Pattern Analysis and Machine Intelligence

|September 4, 2023

Summary

We introduce the Side Adapter Network (SAN), a novel framework for open-vocabulary semantic segmentation. SAN efficiently adapts large vision-language models, achieving state-of-the-art results with significantly reduced computational cost.

More Related Videos

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

Related Experiment Videos

Last Updated: Jul 17, 2025

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

Area of Science:

Computer Vision
Machine Learning
Artificial Intelligence

Background:

Open-vocabulary semantic segmentation enables models to segment arbitrary object categories.
Adapting large-scale vision-language models like CLIP offers a training-economized approach.

Purpose of the Study:

To develop a lightweight and effective framework for open-vocabulary semantic segmentation.
To enhance mainstream segmentation architectures with open-vocabulary capabilities.

Main Methods:

Proposed the Side Adapter Network (SAN) framework, fusing intermediate features from pre-trained CLIP models.
Introduced attention biases via a side adapter network to focus CLIP on informative image regions.
Developed P-SAN and R-SAN variants for pixel-wise and region-wise segmentation.

Main Results:

Achieved state-of-the-art performance on five benchmarks with significantly fewer trainable parameters and GFLOPs.
R-SAN surpassed previous methods by +2.3 mIoU while using only 6% of trainable parameters and <1% of GFLOPs.
Demonstrated the feasibility of transferring R-SAN to video segmentation tasks.

Conclusions:

The SAN framework provides an efficient and effective solution for open-vocabulary semantic segmentation.
SAN empowers existing segmentation models with advanced open-vocabulary capabilities.
The approach shows promise for real-world applications including video segmentation.