Events Temporal Up-sampling

The number of valid events directly affects the performance of event-based tasks, such as reconstruction, detection, and recognition. However, when in low-brightness or slow-moving scenes, events are often sparse and accompanied by noise, which poses challenges for event-based tasks. To solve these challenges, we propose an event temporal up-sampling algorithm to generate more effective and reliable events. Experimental results show that up-sampling events can provide more effective information and improve the performance of downstream tasks, such as improving the quality of reconstructed images and increasing the accuracy of object detection.

For more details, please read our paper "Temporal Up-Sampling for Asynchronous Events".

Introduction

Generate up-sampling events on the correct motion trajectory, which includes estimating the motion trajectory of the events by contrast maximization algorithm and up-sampling the events by the temporal point processes (Hawkes Process for main events, Self-correcting Process for noise).

Code

main.py: up-sampling events

Contrast_Maximization.py: estimate event motion trajectory

Temporal_Point_Processes.py: up-sampling events by Hawkes Process and Self-correcting Process

event_process.py: including warp events, save up-sampling events, show result, etc.

Usage

Change event_path in main.py to your own path.

Dependencies

python=3.8

Publication

If you use this code in an academic context, please cite the following publication:

X. Xiang, L. Zhu, J. Li, Y. Tian and T. Huang, "Temporal Up-Sampling for Asynchronous Events," 2022 IEEE International Conference on Multimedia and Expo (ICME), 2022, pp. 01-06.