Max Robotics — Global Service Robot Directory + US Certification

About

Horia Porav is a researcher focused on making autonomous vehicles and robotic systems robust to adverse visual conditions. His key research areas include visual place recognition, metric localisation, and scene understanding under challenging weather and lighting. Porav’s major contribution is developing methods to transform image appearance—such as converting day to night or summer to winter—using invertible generative models, enabling reliable localisation even when conditions change drastically. He also advanced road scene understanding by improving Inverse Perspective Mapping (IPM) to create more accurate bird’s-eye views, simplifying tasks like lane detection and object tracking. Notably, Porav introduced a clever, low-cost technique for collecting rainy datasets indoors by recording a high-resolution screen, solving the difficult problem of synchronising ground truth with adverse weather. His work, cited over a dozen times across top venues, directly addresses real-world deployment challenges for autonomous systems. Porav’s practical, data-driven approach—from adversarial training to creative dataset generation—has made him a valuable contributor to robust perception in robotics.