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 Experiment Videos

Stereo reconstruction from multiperspective panoramas.

Yin Li1, Heung-Yeung Shum, Chi-Keung Tang

  • 1Computer Science Department, Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong. liyin@ust.hk

IEEE Transactions on Pattern Analysis and Machine Intelligence
|September 24, 2004
PubMed
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

Low-altitude economy: next frontier in spatial exploration and economic development.

National science review·2026
Same author

Occlusion-Aware Instance Segmentation Via BiLayer Network Architectures.

IEEE transactions on pattern analysis and machine intelligence·2023
Same author

GCoNet+: A Stronger Group Collaborative Co-Salient Object Detector.

IEEE transactions on pattern analysis and machine intelligence·2023
Same author

CTRL: Closed-Loop Transcription to an LDR via Minimaxing Rate Reduction.

Entropy (Basel, Switzerland)·2022
Same author

Relative CNN-RNN: Learning Relative Atmospheric Visibility From Images.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2018
Same author

Real-Time Video Stylization Using Object Flows.

IEEE transactions on visualization and computer graphics·2017
Same journal

HardFlow: Hard-Constrained Sampling for Flow-Matching Models Via Trajectory Optimization.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Industrial Brain: Self-Evolving Neuro-Symbolic Autonomy with Causal Resilience for Cyber-Physical Systems.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Adaptive Hardness-Driven Dictionary Distillation for Incomplete Streaming View Clustering.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Mixture of Global and Local Experts with Diffusion Transformer for Controllable Face Generation.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Task-KV: Task-aware KV Cache Optimization via Semantic Differentiation of Attention Heads.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Achieving Text-based Person Retrieval with Any Granularity.

IEEE transactions on pattern analysis and machine intelligence·2026
See all related articles

This paper introduces a novel method for creating 360-degree depth maps using multiperspective panoramas. This approach overcomes limitations of traditional stereo vision, enabling high-quality 3D reconstruction for applications like view interpolation.

Area of Science:

  • Computer Vision
  • 3D Reconstruction
  • Computational Imaging

Background:

  • Traditional stereo matching methods struggle with limited image overlap and complex epipolar geometry.
  • Generating panoramic depth maps requires robust techniques to handle wide fields of view.

Purpose of the Study:

  • To present a new approach for computing panoramic depth maps using multiperspective panoramas.
  • To develop algorithms capable of dense 3D reconstruction from these specialized panoramas.

Main Methods:

  • Constraining camera motion to planar concentric circles to capture images.
  • Resampling perspective images into multiperspective panoramas with uniform sampling in rotation, inverse radial distance, and elevation.
  • Developing two reconstruction algorithms: a cylinder sweep method and a 1D multibaseline matching with tensor voting.

Related Experiment Videos

Main Results:

  • Multiperspective panoramas simplify epipolar geometry to horizontal lines, facilitating stereo matching.
  • The proposed algorithms produce high-quality panoramic depth maps.
  • The method effectively avoids issues associated with limited overlap in conventional multibaseline stereo.

Conclusions:

  • The novel approach enables accurate and dense 3D reconstruction for panoramic scenes.
  • The developed algorithms are efficient and adaptable for various stereo matching techniques.
  • The resulting depth maps are suitable for applications such as view interpolation.