Max Robotics — Global Service Robot Directory + US Certification

About

Mario Bijelic is a researcher specializing in computer vision for adverse weather conditions, neural scene representations, and robust perception systems for autonomous vehicles and robotics. His work addresses one of the most critical challenges in autonomous systems: reliable sensing and imaging through degraded environmental conditions such as fog, rain, and snow. Bijelic's most influential contribution, "ScatterNeRF" (2023, 25 citations), pioneered physically-based inverse neural rendering to see through fog, blending classical scattering physics with modern neural rendering. His earlier "ZeroScatter" work (2021, 14 combined citations) tackled domain transfer for long-distance vision through scattering media, advancing generalizable solutions that require no target-domain supervision. His "SAMFusion" paper (2024, 15 citations) extended this expertise to sensor-adaptive multimodal fusion for 3D object detection under adverse weather, while "Radar Fields" (2024, 11 citations) broke new ground in FMCW radar neural scene reconstruction, complementing existing RGB and LiDAR approaches. Most recently, his introduction of the ML FMEA framework (2025) reflects a growing interest in deploying machine learning safely in safety-critical industries. Collectively, Bijelic's research pushes the boundaries of robust, all-weather perception essential for real-world autonomous systems.