Joint Stereo Video Deblurring, Scene Flow Estimation and Moving Object Segmentation

TL;DR

This paper presents a unified framework for jointly deblurring stereo videos, estimating scene flow, and segmenting moving objects by exploiting their intrinsic connections through a piece-wise planar scene model.

Contribution

It introduces a novel joint approach that simultaneously addresses deblurring, scene flow estimation, and segmentation using a unified energy minimization framework based on a piece-wise planar scene model.

Findings

Significant improvement over state-of-the-art in deblurring stereo videos.

Effective scene flow estimation in complex dynamic scenes.

Accurate moving object segmentation under challenging conditions.

Abstract

Stereo videos for the dynamic scenes often show unpleasant blurred effects due to the camera motion and the multiple moving objects with large depth variations. Given consecutive blurred stereo video frames, we aim to recover the latent clean images, estimate the 3D scene flow and segment the multiple moving objects. These three tasks have been previously addressed separately, which fail to exploit the internal connections among these tasks and cannot achieve optimality. In this paper, we propose to jointly solve these three tasks in a unified framework by exploiting their intrinsic connections. To this end, we represent the dynamic scenes with the piece-wise planar model, which exploits the local structure of the scene and expresses various dynamic scenes. Under our model, these three tasks are naturally connected and expressed as the parameter estimation of 3D scene structure and camera motion (structure and motion for the dynamic scenes). By exploiting the blur model constraint, the moving objects and the 3D scene structure, we reach an energy minimization formulation for joint deblurring, scene flow and segmentation. We evaluate our approach extensively on both synthetic datasets and publicly available real datasets with fast-moving objects, camera motion, uncontrolled lighting conditions and shadows. Experimental results demonstrate that our method can achieve significant improvement in stereo video deblurring, scene flow estimation and moving object segmentation, over state-of-the-art methods.