On-Device Self-Supervised Learning of Low-Latency Monocular Depth from Only Events

Distance between interventions Autonomous Human pilot Self-supervised learning from events + On-device optimizations Events ConvGRU Depth Ego-motion Optical flow Contrast maximization Figure 1 . Online, on-device learning allows robots to "train in their test environment". We improve the time and memory efficiency of the self-supervised contrast maximization pipeline, such that on-board learning of monocular depth from event camera data becomes possible. When deployed on a small drone, online learning leads to better depth estimates and more successful obstacle avoidance behavior.