Learning to Simulate Realistic LiDARs
This work addresses the problem of generating realistic sensor data for autonomous systems, offering an incremental improvement over handcrafted simulation methods.
The paper tackles the challenge of simulating realistic LiDAR sensors for autonomous systems by introducing a data-driven pipeline that learns a mapping from RGB images to LiDAR features like raydrop and intensities from real datasets, resulting in improved vehicle segmentation performance when applied to simulated point clouds.
Simulating realistic sensors is a challenging part in data generation for autonomous systems, often involving carefully handcrafted sensor design, scene properties, and physics modeling. To alleviate this, we introduce a pipeline for data-driven simulation of a realistic LiDAR sensor. We propose a model that learns a mapping between RGB images and corresponding LiDAR features such as raydrop or per-point intensities directly from real datasets. We show that our model can learn to encode realistic effects such as dropped points on transparent surfaces or high intensity returns on reflective materials. When applied to naively raycasted point clouds provided by off-the-shelf simulator software, our model enhances the data by predicting intensities and removing points based on the scene's appearance to match a real LiDAR sensor. We use our technique to learn models of two distinct LiDAR sensors and use them to improve simulated LiDAR data accordingly. Through a sample task of vehicle segmentation, we show that enhancing simulated point clouds with our technique improves downstream task performance.