A hybrid motion estimation technique for fisheye video sequences based on equisolid re-projection

TL;DR

This paper presents a hybrid motion estimation method tailored for fisheye videos that improves accuracy by using equisolid re-projection, combining perspective domain search with traditional methods, leading to notable PSNR gains.

Contribution

It introduces a novel motion estimation approach that accounts for fisheye radial distortion using equisolid re-projection, enhancing performance over conventional methods.

Findings

Up to 1.14 dB PSNR improvement on synthetic sequences

Up to 0.96 dB PSNR improvement on real-world data

Maximum gains of 2.40 dB and 1.39 dB for selected frames

Abstract

Capturing large fields of view with only one camera is an important aspect in surveillance and automotive applications, but the wide-angle fisheye imagery thus obtained exhibits very special characteristics that may not be very well suited for typical image and video processing methods such as motion estimation. This paper introduces a motion estimation method that adapts to the typical radial characteristics of fisheye video sequences by making use of an equisolid re-projection after moving part of the motion vector search into the perspective domain via a corresponding back-projection. By combining this approach with conventional translational motion estimation and compensation, average gains in luminance PSNR of up to 1.14 dB are achieved for synthetic fish-eye sequences and up to 0.96 dB for real-world data. Maximum gains for selected frame pairs amount to 2.40 dB and 1.39 dB for synthetic and real-world data, respectively.