Helena Kotthaus

Papers

Total Citations

H-Index

About

Helena Kotthaus is a leading researcher at the intersection of explainable artificial intelligence (XAI) and 3D computer vision, with a particular focus on point cloud neural networks. Her major contribution lies in pioneering surrogate model-based explainability methods tailored for point cloud data, a critical area for autonomous driving and robotics where 3D sensors are ubiquitous. Her most-cited work (2022, 25 citations) addresses a significant gap: while point cloud neural networks excel in real-time performance, their "black-box" nature hinders trust and deployment. Kotthaus’s approach provides interpretable, post-hoc explanations for these complex models, enabling engineers and regulators to understand decision-making processes in safety-critical applications. Her earlier work (2021, 3 citations) laid the foundation for this methodology, demonstrating its feasibility. By bridging the gap between high-performance 3D perception and model transparency, Kotthaus has made a pivotal impact on trustworthy AI in autonomous systems. Her research is essential reading for students and practitioners seeking to build robust, explainable AI for real-world 3D environments.