Reliability of GAN Generated Data to Train and Validate Perception Systems for Autonomous Vehicles

Weihuang Xu, Nasim Souly, Pratik Prabhanjan Brahma; Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) Workshops, 2021, pp. 171-180

Abstract


Autonomous systems deployed in the real world have to deal with potential problem causing situations that they have never seen during their training phases. Due to the long tail nature of events, it is difficult to collect large amount of data for such corner cases. While simulation is one plausible solution, recent developments in the field of Generative Adversarial Networks (GANs) have shown how they can be used to generate and augment realistic data without exhibiting a domain shift from actual real data. In this manuscript, we empirically analyze and propose novel solutions for the trust that we can place on GAN generated data for training and validation of vision based perception modules like object detection and scenario classification.

Related Material


[pdf] [supp]
[bibtex]
@InProceedings{Xu_2021_WACV, author = {Xu, Weihuang and Souly, Nasim and Brahma, Pratik Prabhanjan}, title = {Reliability of GAN Generated Data to Train and Validate Perception Systems for Autonomous Vehicles}, booktitle = {Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) Workshops}, month = {January}, year = {2021}, pages = {171-180} }