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: Mar 24, 2026

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

Human Parsing with Contextualized Convolutional Neural Network.

Xiaodan Liang, Chunyan Xu, Xiaohui Shen

IEEE Transactions on Pattern Analysis and Machine Intelligence

|March 9, 2016

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

Symptom Clusters and Quality of Life in Cervical Cancer Patients During the Perioperative Period: A Longitudinal Study.

Patient preference and adherence·2026

Same author

MonSter++: Unified Stereo Matching, Multi-View Stereo, and Real-Time Stereo With Monodepth Priors.

IEEE transactions on pattern analysis and machine intelligence·2026

Same author

Perioperative Symptom Trajectories and a Risk Prediction Model for Cervical Cancer: A Prospective Longitudinal Study.

International journal of women's health·2026

Same author

Measuring integrated hypertension care in community health centers: a cross-sectional study based on the Rainbow Model of Integrated Care in Chengdu, China.

BMC health services research·2026

Same author

Lamprey 3D single-cell transcriptomics reveals ancestral and specialized features of the vertebrate brain.

Science (New York, N.Y.)·2026

Same author

p53 overexpression counteracts the pro-survival effect of Bcl-2 by restoring MAMs function in cutaneous squamous cell carcinoma.

Cell division·2026

A new Contextualized Convolutional Neural Network (Co-CNN) significantly improves human parsing accuracy. This deep learning model integrates multiple contextual cues for precise pixelwise categorization, outperforming existing methods.

Area of Science:

Computer Vision
Deep Learning
Image Analysis

Background:

Human parsing, the task of pixelwise categorization of human body parts, is crucial for various applications.
Existing methods often struggle to effectively integrate diverse contextual information for accurate parsing.

Purpose of the Study:

To introduce a novel Contextualized Convolutional Neural Network (Co-CNN) architecture for enhanced human parsing.
To effectively integrate cross-layer, global image-level, semantic edge, and super-pixel contexts within a unified network.

Main Methods:

Developed a Contextualized Convolutional Neural Network (Co-CNN) integrating multiple contextual information sources.
Employed a local-to-global-to-local structure for cross-layer context, auxiliary global image-level label prediction, semantic edge context integration, and within/cross-super-pixel smoothing/voting for local consistency.

More Related Videos

Decoding Natural Behavior from Neuroethological Embedding

Decoding Natural Behavior from Neuroethological Embedding

Published on: October 3, 2025

Related Experiment Videos

Last Updated: Mar 24, 2026

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

Decoding Natural Behavior from Neuroethological Embedding

Decoding Natural Behavior from Neuroethological Embedding

Published on: October 3, 2025

Utilized end-to-end pixelwise categorization for human parsing.

Main Results:

Co-CNN achieved a superior F-1 score of 81.72% on a large public dataset, significantly outperforming state-of-the-art methods like M-CNN (62.81%) and ATR (64.38%).
Training with a newly collected large dataset further boosted the F-1 score to 85.36%.
Demonstrated significant superiority over existing methods in comprehensive evaluations.

Conclusions:

The proposed Co-CNN architecture effectively integrates diverse contextual information for state-of-the-art human parsing performance.
The model's ability to leverage multi-level context leads to significant improvements in pixelwise categorization accuracy.
Co-CNN represents a substantial advancement in the field of human parsing.