Robust Fundamental Matrix Estimation from Single Image Motion Blur

2026-05-02

AI Summary

This work addresses the challenge of establishing reliable feature correspondences in single motion-blurred images caused by camera motion, where conventional methods fail. It proposes the first approach to estimate the fundamental matrix directly from a single blurred frame by modeling the implicit temporal correspondences encoded in the blur trajectory across the exposure window. The method integrates an enhanced fundamental matrix estimation algorithm with a RANSAC sampling strategy that leverages uncertainty predictions derived from the smear pattern to effectively resolve temporal ambiguity. Experiments demonstrate that the proposed approach robustly recovers the fundamental matrix encoding 3D camera motion on both synthetic and real-world motion-blurred datasets, and its effectiveness is further validated through downstream tasks such as motion segmentation.

Abstract

In this paper, we introduce a challenging task: extracting a fundamental matrix from a single motion blurred image. For a camera moving in 3D during exposure, the smear paths in the blurry image contain cues and constraints on this motion. We demonstrate the feasibility of establishing correspondences between two time instances within the camera exposure window, and that these can be used to robustly infer a fundamental matrix, which summarizes the motion of the camera during the exposure time. The inferred fundamental matrix is unique up to a transpose, corresponding to an ambiguity of the direction of time. Due to this per-smear ambiguity, classic methods, such as the 8-point algorithm, are no longer usable. The proposed method modifies the estimation to work on time-direction ambiguous correspondences. To improve the robustness of the fundamental matrix estimation, we also propose to incorporate an uncertainty measurement in smear pattern prediction and use it in the sampling process of the estimator. Experiments on synthetic and real-world motion-blur datasets demonstrate that our approach is able to estimate the fundamental matrix encoding the 3D camera motion, from single frames. Practical applicability is demonstrated on the downstream task of motion segmentation.

Problem

• Fundamental Matrix
• Motion Blur
• Single Image
• Camera Motion
• 3D Motion

Innovation

• fundamental matrix
• motion blur
• single-image motion estimation
• time-direction ambiguity
• robust estimation