Single Photon LiDAR Compression: An Overview

Real-time mapping of the surroundings plays a pivotal role in ensuring the success of autonomous vehicles. Among the technologies showing promise for depth mapping is Single Photon Light Detection and Ranging (SPL) which offers impressive long-range capabilities, superior depth resolution, and energy-efficient laser sources. However, the adoption of high-resolution, high frame-rate SPL systems results in significant data volumes, often reaching several Gbit/s. This creates a throughput bottleneck that impedes real-time applications and necessitates the development of effective compression schemes. This paper addresses the challenges associated with SPL compression and provides a comprehensive overview of recent advancements in this domain. The review encompasses various compression methods, including algorithmic and/or hardware-based approaches, spatial and/or temporal compression techniques. We also discuss future directions, such as integrating deep learning and fusion techniques with other sensing modalities.