Research

Different outdoor illumination conditions drastically alter the appearance of urban scenes, and can harm the performance of image-based robot perception systems if they have not been seen during training. Towards this goal, we propose LightSim, a neural lighting camera simulation system that enables diverse, realistic, and controllable data generation. LightSim automatically builds relightable digital twins at scale from collected raw sensor data…

We use arithmetic codes to create a static representation of AVL trees—a class of self-balancing binary search trees—using less than 0.99933 bits per node. Our encoding supports efficient full traversal of the tree. To evaluate how close to optimal our encoding is, we derive asymptotics for the information theoretic lower-bound on the number of bits needed to store AVL trees…

As self-driving systems become better, simulating scenarios where the autonomy stack may fail becomes more important. Traditionally, those scenarios are generated for a few scenes with respect to the planning module that takes ground-truth actor states as input. This does not scale and cannot identify all possible autonomy failures, such as perception failures due to occlusion. In this paper, we propose AdvSim, an adversarial framework to generate safety-critical scenarios…