AI Navigate

Single-View Rolling-Shutter SfM

arXiv cs.CV / 3/13/2026

📰 NewsIdeas & Deep AnalysisModels & Research

Key Points

  • The article tackles single-view rolling-shutter (RS) structure-from-motion (SfM) and develops a geometry characterization for observed world points and lines.
  • It systematically derives which motion and scene parameters can be recovered from a single RS image and outlines minimal reconstruction problems.
  • The authors present proof-of-concept solvers and evaluate several representative cases to assess feasibility and practical limitations.
  • The work highlights the potential of RS-SfM for ubiquitous cameras while clarifying remaining challenges and boundary conditions.

Abstract

Rolling-shutter (RS) cameras are ubiquitous, but RS SfM (structure-from-motion) has not been fully solved yet. This work suggests an approach to remedy this: We characterize RS single-view geometry of observed world points or lines. Exploiting this geometry, we describe which motion and scene parameters can be recovered from a single RS image and systematically derive minimal reconstruction problems. We evaluate several representative cases with proof-of-concept solvers, highlighting both feasibility and practical limitations.