UnRectDepthNet: Self-Supervised Monocular Depth Estimation using a Generic Framework for Handling Common Camera Distortion Models

TL;DR

UnRectDepthNet introduces a self-supervised monocular depth estimation framework that effectively handles unrectified, distorted camera images, including fisheye lenses, without the need for rectification, maintaining accuracy and expanding applicability.

Contribution

The paper presents a novel generic framework that implicitly learns distortion models within a CNN for depth estimation from unrectified videos, eliminating the need for rectification and broadening the scope to wide-angle cameras.

Findings

Achieves state-of-the-art results on KITTI datasets.

Performs well on wide-angle fisheye cameras with 190° FOV.

Maintains accuracy comparable to rectified methods without rectification costs.

Abstract

In classical computer vision, rectification is an integral part of multi-view depth estimation. It typically includes epipolar rectification and lens distortion correction. This process simplifies the depth estimation significantly, and thus it has been adopted in CNN approaches. However, rectification has several side effects, including a reduced field of view (FOV), resampling distortion, and sensitivity to calibration errors. The effects are particularly pronounced in case of significant distortion (e.g., wide-angle fisheye cameras). In this paper, we propose a generic scale-aware self-supervised pipeline for estimating depth, euclidean distance, and visual odometry from unrectified monocular videos. We demonstrate a similar level of precision on the unrectified KITTI dataset with barrel distortion comparable to the rectified KITTI dataset. The intuition being that the rectification step can be implicitly absorbed within the CNN model, which learns the distortion model without increasing complexity. Our approach does not suffer from a reduced field of view and avoids computational costs for rectification at inference time. To further illustrate the general applicability of the proposed framework, we apply it to wide-angle fisheye cameras with 190$^\circ$ horizontal field of view. The training framework UnRectDepthNet takes in the camera distortion model as an argument and adapts projection and unprojection functions accordingly. The proposed algorithm is evaluated further on the KITTI rectified dataset, and we achieve state-of-the-art results that improve upon our previous work FisheyeDistanceNet. Qualitative results on a distorted test scene video sequence indicate excellent performance https://youtu.be/K6pbx3bU4Ss.