Generative Adversarial Synthesis of Radar Point Cloud Scenes
This work addresses the need for efficient dataset creation in automotive radar testing, though it is incremental as it applies existing GAN methods to a specific domain.
The paper tackles the problem of generating realistic radar point cloud scenes for automotive radar validation by using a GAN model, achieving a performance of ~87% similarity to real scenes.
For the validation and verification of automotive radars, datasets of realistic traffic scenarios are required, which, how ever, are laborious to acquire. In this paper, we introduce radar scene synthesis using GANs as an alternative to the real dataset acquisition and simulation-based approaches. We train a PointNet++ based GAN model to generate realistic radar point cloud scenes and use a binary classifier to evaluate the performance of scenes generated using this model against a test set of real scenes. We demonstrate that our GAN model achieves similar performance (~87%) to the real scenes test set.