DurLAR: A High-Fidelity 128-Channel LiDAR Dataset with Panoramic Ambient and Reflectivity Imagery for Multi-Modal Autonomous Driving Applications

Abstract

We present DurLAR, a high-fidelity 128-channel 3D Li-DAR dataset with panoramic ambient (near infrared) and reflectivity imagery, as well as a sample benchmark task using depth estimation for autonomous driving applications. Our driving platform is equipped with a high resolution 128 channel LiDAR, a 2MPix stereo camera, a lux meter and a GNSS/INS system. Ambient and reflectivity images are made available along with the LiDAR point clouds to facilitate multi-modal use of concurrent ambient and reflectivity scene information. Leveraging DurLAR, with a resolution exceeding that of prior benchmarks, we consider the task of monocular depth estimation and use this increased availability of higher resolution, yet sparse ground truth scene depth information to propose a novel joint supervised/selfsupervised loss formulation. We compare performance over both our new DurLAR dataset, the established KITTI benchmark and the Cityscapes dataset. Our evaluation shows our joint use supervised and self-supervised loss terms, enabled via the superior ground truth resolution and availability within DurLAR improves the quantitative and qualitative performance of leading contemporary monocular depth estimation approaches (RM SE = 3.639, SqRel = 0.936).

3-D Scene Flow Estimation on Pseudo-LiDAR: Bridging the Gap on Estimating Point Motion

“…The accuracy of learning 3D scene flow on the 128-beam LiDAR signals [14] is improved compared to the 64-beam LiDAR signals. According to the sentence according to the quantitative results of Table I and Table II, the proposed framework for learning 3D scene flow from pseudo-LiDAR signals still presents greater advantages.…”

“…To further demonstrate the denseness advantage of the pseudo-LiDAR point cloud proposed in section III-C1, the scene flow estimator is trained on a denser LiDAR point clouds from the high-fidelity 128-Channel LiDAR Dataset (DurLAR) [14]. To be fair, we perform the same processing as PointPWC-Net [9] for the LiDAR point cloud in DurLAR.…”

“…From the bottom of Table I, the 3D scene flow estimator is trained from scratch on a synthesized dataset [12], Li-DAR point clouds [14], [20], and real-world stereo/monocular images [20], respectively. It is obvious that self-supervised learning works best on pseudo-LiDAR point clouds estimated from images.…”

“…Although the cost of LiDAR is shrinking year by year, there is still an undeniable cost difference between LiDAR and camera. Recently, Ouster released an inexpensive OS1-128 LiDAR, but the price still costs $18,000 [14]. The stereo camera system may provide a low-cost backup system for the LiDAR-based scheme of scene flow estimation.…”

On Depth Error from Spherical Camera Calibration within Omnidirectional Stereo Vision

“…Whilst the cost of alternative technologies for 3D scene depth recovery, such as LiDAR [4], remain prohibitively high for many applications, stereo vision based sensing offers a good compromise between accuracy and cost and is used widely in robotics [5], [6], [7], [8], object recognition [9], [10], and 3D scene reconstruction [11], [12]. However, a key limitation of common stereo vision solutions is the limited Field of View (FoV) afforded by the use of conventional cameras.…”