Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

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

Selenium Doping and Vacancy Engineering to Regulate the P-Band Center of Sulfur in Zinc Sulfide for High-Performance Sodium-Ion Storage.

Nano letters·2026
Same author

A Biomimetic Bidirectional Interphase Enabled by a Single Molecule for Ultra-Stable Zn-I<sub>2</sub> Batteries.

Small (Weinheim an der Bergstrasse, Germany)·2026
Same author

A Spatiotemporally Coupled Carbon Flux Monitoring System for Salt Marsh Wetlands Based on Integrated Land-Air Collaborative Intelligence.

Sensors (Basel, Switzerland)·2026
Same author

An adaptive bidirectional mapping framework for the customized design of interpenetrating phase composite scaffolds.

Computer methods in biomechanics and biomedical engineering·2026
Same author

Sustainable Cellulose Enables Blue Energy Toward Osmotic Energy Conversion.

Nano-micro letters·2026
Same author

S-atom dislocation-induced room-temperature ferroelectricity in two-dimensional α-MnS semiconductor.

Nature communications·2026

Related Experiment Video

Updated: Dec 26, 2025

Collecting and Processing Drone-based Remotely Sensed Data for Use in Forest Recovery Monitoring
08:16

Collecting and Processing Drone-based Remotely Sensed Data for Use in Forest Recovery Monitoring

Published on: October 24, 2025

358

Outdoor RGBD Instance Segmentation with Residual Regretting Learning.

Zhengtian Xu, Shu Liu, Jianping Shi

    IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
    |March 7, 2020
    PubMed
    Summary

    This study introduces a residual regretting mechanism to improve outdoor instance segmentation using RGBD data. The novel approach enhances depth map utilization, achieving state-of-the-art results on challenging datasets.

    More Related Videos

    Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
    04:48

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

    Published on: November 30, 2022

    3.3K
    Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
    03:31

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

    Published on: December 15, 2023

    938

    Related Experiment Videos

    Last Updated: Dec 26, 2025

    Collecting and Processing Drone-based Remotely Sensed Data for Use in Forest Recovery Monitoring
    08:16

    Collecting and Processing Drone-based Remotely Sensed Data for Use in Forest Recovery Monitoring

    Published on: October 24, 2025

    358
    Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
    04:48

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

    Published on: November 30, 2022

    3.3K
    Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
    03:31

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

    Published on: December 15, 2023

    938

    Area of Science:

    • Computer Vision
    • Robotics
    • Artificial Intelligence

    Background:

    • Indoor semantic segmentation with RGBD input shows progress.
    • Outdoor instance segmentation faces challenges due to ambiguous depth maps.

    Purpose of the Study:

    • To address challenges in outdoor instance segmentation using RGBD data.
    • To improve the utilization of depth map information for object detection and segmentation.

    Main Methods:

    • Proposed a residual regretting mechanism integrated into the Mask R-CNN framework.
    • Developed a regretting cascade to refine depth map information.
    • Utilized a novel residual connection to robustly combine RGB and depth data.

    Main Results:

    • Demonstrated effectiveness on the Cityscapes and KITTI datasets.
    • Achieved state-of-the-art performance in RGBD instance segmentation.
    • Showcased a 13.4% relative improvement over standard Mask R-CNN on Cityscapes using depth cues.

    Conclusions:

    • The residual regretting mechanism significantly enhances outdoor instance segmentation.
    • This method effectively handles ambiguous outdoor depth maps.
    • The approach offers a robust solution for RGBD instance segmentation in real-world scenarios.