Constructing and Evaluating Complex Event-based Datasets for Increasing Performance of Instance Segmentation Models

Constructing and Evaluating Complex Event-based Datasets for Increasing Performance of Instance Segmentation Models

Manolache, Alexandru-Dragos (TU Delft Electrical Engineering, Mathematics and Computer Science; TU Delft Pattern Recognition and Bioinformatics)

Tömen, N. (mentor)
Strafforello, O. (mentor)
Liu, X. (mentor)
Cavalcante Siebert, L. (graduation committee)

Delft University of Technology

Computer Science and Engineering

CSE3000 Research Project

2022-06-22

Event-based cameras represent a new alternative to traditional frame based sensors, with advantages in lower output bandwidth, lower latency and higher dynamic range, thanks to their independent, asynchronous pixels. These advantages prompted the development of computer vision methods on event data in the last decade, however event-based datasets are still in early stages in terms of size and complexity compared to normal datasets (e.g. ImageNet). This paper explores event data augmentation by superimposing two existing event datasets (N-MNIST and N-Caltech101) and by adding uniform noise. It shows that training an instance segmentation model on noisy datasets does not improve its performance, but the amount and type of noise added in the background decreases the performance of such model.

Event-based vision
Event-based datasets
Event-based data augmentation
Dynamic Vision Sensor

http://resolver.tudelft.nl/uuid:fd7d1ed1-dad1-4c3d-99c5-633ce2764460

Student theses

bachelor thesis

© 2022 Alexandru-Dragos Manolache