Cityscapes 3D: Dataset and Benchmark for 9 DoF Vehicle Detection

TL;DR

This paper introduces Cityscapes 3D, a new dataset with stereo RGB-based 3D vehicle annotations for autonomous driving, enabling improved 3D detection and benchmarking without lidar data.

Contribution

The paper presents a novel 3D vehicle dataset derived from stereo RGB images, with accurate 3D annotations capturing all nine degrees of freedom, and provides benchmarks for 3D detection tasks.

Findings

Stereo RGB annotations achieve high accuracy in 3D vehicle localization.

The dataset enables fair comparison of 3D detection methods.

Benchmark results highlight the challenges and progress in monocular 3D detection.

Abstract

Detecting vehicles and representing their position and orientation in the three dimensional space is a key technology for autonomous driving. Recently, methods for 3D vehicle detection solely based on monocular RGB images gained popularity. In order to facilitate this task as well as to compare and drive state-of-the-art methods, several new datasets and benchmarks have been published. Ground truth annotations of vehicles are usually obtained using lidar point clouds, which often induces errors due to imperfect calibration or synchronization between both sensors. To this end, we propose Cityscapes 3D, extending the original Cityscapes dataset with 3D bounding box annotations for all types of vehicles. In contrast to existing datasets, our 3D annotations were labeled using stereo RGB images only and capture all nine degrees of freedom. This leads to a pixel-accurate reprojection in the RGB image and a higher range of annotations compared to lidar-based approaches. In order to ease multitask learning, we provide a pairing of 2D instance segments with 3D bounding boxes. In addition, we complement the Cityscapes benchmark suite with 3D vehicle detection based on the new annotations as well as metrics presented in this work. Dataset and benchmark are available online.