SVDistNet: Self-Supervised Near-Field Distance Estimation on Surround View Fisheye Cameras

TL;DR

SVDistNet introduces a self-supervised, camera-geometry adaptive approach for near-field distance estimation using surround-view fisheye cameras, improving generalization and accuracy in autonomous driving scenarios.

Contribution

The paper proposes novel camera-geometry adaptive multi-scale convolutions and vector-based self-attention networks for robust, generalizable depth estimation across diverse fisheye camera setups.

Findings

Significant improvement over previous methods on Fisheye WoodScape dataset.

Achieves state-of-the-art results on self-supervised monocular distance estimation.

Demonstrates strong generalization across different camera angles and datasets.

Abstract

A 360{\deg} perception of scene geometry is essential for automated driving, notably for parking and urban driving scenarios. Typically, it is achieved using surround-view fisheye cameras, focusing on the near-field area around the vehicle. The majority of current depth estimation approaches focus on employing just a single camera, which cannot be straightforwardly generalized to multiple cameras. The depth estimation model must be tested on a variety of cameras equipped to millions of cars with varying camera geometries. Even within a single car, intrinsics vary due to manufacturing tolerances. Deep learning models are sensitive to these changes, and it is practically infeasible to train and test on each camera variant. As a result, we present novel camera-geometry adaptive multi-scale convolutions which utilize the camera parameters as a conditional input, enabling the model to generalize to previously unseen fisheye cameras. Additionally, we improve the distance estimation by pairwise and patchwise vector-based self-attention encoder networks. We evaluate our approach on the Fisheye WoodScape surround-view dataset, significantly improving over previous approaches. We also show a generalization of our approach across different camera viewing angles and perform extensive experiments to support our contributions. To enable comparison with other approaches, we evaluate the front camera data on the KITTI dataset (pinhole camera images) and achieve state-of-the-art performance among self-supervised monocular methods. An overview video with qualitative results is provided at https://youtu.be/bmX0UcU9wtA. Baseline code and dataset will be made public.