# Talk Proposal: Towards the Realistic Evaluation of Evasion Attacks using CARLA

arXiv: https://arxiv.org/abs/1904.12622 · HTML: https://mlsanity.com/paper/1904.12622/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 18 Apr 2019.

## Built on by (most cited first, 2 of 2)

- [DriveGuard: Robustification of Automated Driving Systems with Deep Spatio-Temporal Convolutional Autoencoder](https://mlsanity.com/paper/2111.03480/) (2021)
- [Synthetic Dataset Generation for Adversarial Machine Learning Research](https://mlsanity.com/paper/2207.10719/) (2022)

## Abstract (arXiv)

In this talk we describe our content-preserving attack on object detectors, ShapeShifter, and demonstrate how to evaluate this threat in realistic scenarios. We describe how we use CARLA, a realistic urban driving simulator, to create these scenarios, and how we use ShapeShifter to generate content-preserving attacks against those scenarios.