Low Complexity Learning-based Lossless Event-based Compression

Abstract

—Event cameras are a cutting-edge type of visual sensors that capture data by detecting brightness changes at the pixel level asynchronously. These cameras offer numerous benefits over conventional cameras, including high temporal resolution, wide dynamic range, low latency, and lower power consumption. However, the substantial data rates they produce require efficient compression techniques, while also fulfilling other typical application requirements, such as the ability to respond to visual changes in real-time or near real-time. Additionally, many eventbased applications demand high accuracy, making lossless coding desirable, as it retains the full detail of the sensor data. Learningbased methods show great potential due to their ability to model the unique characteristics of event data thus allowing to achieve high compression rates. This paper proposes a low-complexity lossless coding solution based on the quadtree representation that outperforms traditional compression algorithms in efficiency and speed, ensuring low computational complexity and minimal delay for real-time applications. Experimental results show that the proposed method delivers better compression ratios, i.e., with fewer bits per event, and lower computational complexity compared to current lossless data compression methods.