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

Parallel Processing01:20

Parallel Processing

350
The brain processes sensory information rapidly due to parallel processing, which involves sending data across multiple neural pathways at the same time. This method allows the brain to manage various sensory qualities, such as shapes, colors, movements, and locations, all concurrently. For instance, when observing a forest landscape, the brain simultaneously processes the movement of leaves, the shapes of trees, the depth between them, and the various shades of green. This enables a quick and...
350
Depth Perception and Spatial Vision01:15

Depth Perception and Spatial Vision

1.1K
Depth perception is the ability to perceive objects three-dimensionally. It relies on two types of cues: binocular and monocular. Binocular cues depend on the combination of images from both eyes and how the eyes work together. Since the eyes are in slightly different positions, each eye captures a slightly different image. This disparity between images, known as binocular disparity, helps the brain interpret depth. When the brain compares these images, it determines the distance to an object.
1.1K

You might also read

Related Articles

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

Sort by
Same author

Pure FPGA Implementation of an HOG Based Real-Time Pedestrian Detection System.

Sensors (Basel, Switzerland)·2018
Same author

Warfarin Use in Hemodialysis Patients With Atrial Fibrillation: A Systematic Review of Stroke and Bleeding Outcomes.

Canadian journal of kidney health and disease·2017
Same author

Interfacially polymerized layers for oxygen enrichment: a method to overcome Robeson's upper-bound limit.

ACS applied materials & interfaces·2013
See all related articles

Related Experiment Video

Updated: Oct 10, 2025

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

677

Real-Time Semantic Segmentation with Dual Encoder and Self-Attention Mechanism for Autonomous Driving.

Yu-Bang Chang1, Chieh Tsai1, Chang-Hong Lin1

  • 1Department of Electronic and Computer Engineering, National Taiwan University of Science and Technology, Taipei City 106, Taiwan.

Sensors (Basel, Switzerland)
|December 10, 2021
PubMed
Summary
This summary is machine-generated.

This study introduces a novel deep learning architecture for real-time semantic segmentation in autonomous driving. The proposed dual encoder and self-attention model significantly improves accuracy while maintaining high inference speeds for edge devices.

Keywords:
autonomous drivingconvolution neural networkdeep learningedge devicesimage recognitionreal-time semantic segmentation

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

3.0K

Related Experiment Videos

Last Updated: Oct 10, 2025

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

677
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

3.0K

Area of Science:

  • Computer Vision
  • Deep Learning
  • Autonomous Driving Systems

Background:

  • Real-time semantic segmentation is crucial for autonomous driving.
  • Existing methods struggle to balance accuracy and inference speed on edge devices.
  • A significant accuracy gap persists between real-time and general semantic segmentation models.

Purpose of the Study:

  • To develop a deep learning architecture for real-time semantic segmentation with an optimal accuracy-inference time trade-off.
  • To enable efficient deployment of semantic segmentation models on vehicle-integrated edge devices.
  • To bridge the performance gap between real-time and general semantic segmentation.

Main Methods:

  • Proposed a novel network architecture utilizing a dual encoder.
  • Incorporated a self-attention mechanism to enhance feature representation.
  • Evaluated the model on high-resolution datasets for autonomous driving scenarios.

Main Results:

  • Achieved 78.6% mean Intersection over Union (mIoU) accuracy.
  • Reached an inference speed of 39.4 Frames Per Second (FPS).
  • Demonstrated performance at a 1024 × 2048 resolution on the Cityscapes dataset.

Conclusions:

  • The proposed dual encoder and self-attention network offers a superior balance of accuracy and speed for real-time semantic segmentation.
  • This architecture is well-suited for deployment on edge devices in autonomous vehicles.
  • The findings represent a significant advancement in real-time semantic segmentation for autonomous driving applications.