Eliminating the Blind Spot: Adapting 3D Object Detection and Monocular Depth Estimation to 360{\deg} Panoramic Imagery

TL;DR

This paper develops a method to adapt existing 3D object detection and depth estimation models for 360-degree panoramic images, enabling surround sensing in autonomous vehicles without requiring panoramic training data.

Contribution

It introduces a style and projection transformation approach to adapt conventional models to panoramic imagery, allowing monocular depth and 3D pose estimation without panoramic labels.

Findings

Successful adaptation of models to panoramic images

Qualitative evaluation on crowd-sourced data

Quantitative benchmarking in automotive simulation

Abstract

Recent automotive vision work has focused almost exclusively on processing forward-facing cameras. However, future autonomous vehicles will not be viable without a more comprehensive surround sensing, akin to a human driver, as can be provided by 360{\deg} panoramic cameras. We present an approach to adapt contemporary deep network architectures developed on conventional rectilinear imagery to work on equirectangular 360{\deg} panoramic imagery. To address the lack of annotated panoramic automotive datasets availability, we adapt a contemporary automotive dataset, via style and projection transformations, to facilitate the cross-domain retraining of contemporary algorithms for panoramic imagery. Following this approach we retrain and adapt existing architectures to recover scene depth and 3D pose of vehicles from monocular panoramic imagery without any panoramic training labels or calibration parameters. Our approach is evaluated qualitatively on crowd-sourced panoramic images and quantitatively using an automotive environment simulator to provide the first benchmark for such techniques within panoramic imagery.