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

Multi-input and Multi-variable systems

Multi-input and Multi-variable systems

Cruise control systems in cars are designed as multi-input systems to maintain a driver's desired speed while compensating for external disturbances such as changes in terrain. The block diagram for a cruise control system typically includes two main inputs: the desired speed set by the driver and any external disturbances, such as the incline of the road. By adjusting the engine throttle, the system maintains the vehicle's speed as close to the desired value as possible.
In the absence of...

You might also read

Related Articles

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

Sort by

Same author

The role of cGAS-STING pathway in the development of radiation-induced lung injury.

Journal of cancer research and clinical oncology·2025

Same author

Microbial manganese redox cycling drives co-removal of nitrate and ammonium.

Journal of environmental management·2025

Same author

Atomically Dispersed Ta-O-Co Sites Capable of Mitigating Side Reaction Occurrence for Stable Lithium-Oxygen Batteries.

Journal of the American Chemical Society·2025

Same author

Cortical Gyrification and Cognitive Decline in the Human Brain With Type 2 Diabetes Mellitus.

Brain and behavior·2025

Same author

Fluorine Doping-Assisted Reconstruction of Isolated Cu Sites for CO<sub>2</sub> Electroreduction Toward Multicarbon Products.

Advanced materials (Deerfield Beach, Fla.)·2025

Same author

Repurposing of phosphodiesterase-5 inhibitor sildenafil as a therapeutic agent to prevent gastric cancer growth through suppressing c-MYC stability for IL-6 transcription.

Communications biology·2025

Same journal

A Comprehensive Survey on Multimodal Recommender Systems: Taxonomy, Evaluation, and Future Directions.

IEEE transactions on pattern analysis and machine intelligence·2026

Same journal

Benchmarking the Robustness of Autonomous Driving to Environmental Illusions: A Lane Perception Perspective.

IEEE transactions on pattern analysis and machine intelligence·2026

Same journal

Learning Topology-Aware Representations via Test-Time Adaptation for Anomaly Segmentation.

IEEE transactions on pattern analysis and machine intelligence·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

See all related articles

Search research articles

Related Experiment Video

Updated: Apr 15, 2026

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data

Published on: August 13, 2014

Uni-to-Multi Modal Knowledge Distillation for Bidirectional LiDAR-Camera Semantic Segmentation.

Tianfang Sun, Zhizhong Zhang, Xin Tan

IEEE Transactions on Pattern Analysis and Machine Intelligence

|August 29, 2024

Summary

This summary is machine-generated.

This study introduces a novel approach for robust semantic segmentation by fusing LiDAR and image data. The method effectively addresses modality alignment challenges, improving performance on autonomous driving datasets.

More Related Videos

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Published on: May 7, 2019

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

Related Experiment Videos

Last Updated: Apr 15, 2026

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data

Published on: August 13, 2014

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Published on: May 7, 2019

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

Area of Science:

Computer Vision
Robotics
Machine Learning

Background:

Combining LiDAR and camera data for semantic segmentation offers significant potential in autonomous systems.
Heterogeneity between LiDAR (point clouds) and images (pixels) presents modality alignment challenges, hindering cross-modal fusion.
Existing methods struggle with projected points outside image planes and limited data augmentation due to geometric inconsistencies.

Purpose of the Study:

To develop a robust cross-modal approach for semantic segmentation that overcomes modality alignment issues.
To enable reliable predictions even with missing sensor data (e.g., images).
To improve the effectiveness of training cross-modal networks using data augmentation.

Main Methods:

A bidirectional feature fusion strategy is proposed to simultaneously impute missing image features and perform cross-modal fusion.
A Uni-to-Multi modal Knowledge Distillation (U2MKD) framework transfers knowledge from single-modality teachers to a cross-modality student network.
The approach addresses augmentation misalignment and enhances student network training.

Main Results:

Experiments on nuScenes, Waymo, and SemanticKITTI datasets validate the proposed method's effectiveness.
The approach achieved an 8.3 mIoU gain over LiDAR-only baselines on the nuScenes validation set.
State-of-the-art performance was attained across all three benchmark datasets.

Conclusions:

The proposed cross-modal knowledge imputation and transition approach effectively solves modality alignment problems in sensor fusion for semantic segmentation.
The bidirectional feature fusion and U2MKD framework significantly enhance segmentation accuracy and robustness, particularly in challenging real-world scenarios.
This work advances the capabilities of autonomous systems by enabling more reliable and accurate environmental perception through improved sensor fusion techniques.