Maksim Liubimov

Title: Maksim Liubimov: Advancing Camera-Based Perception for Autonomous Vehicles.

Introduction

Maksim Liubimov is an inventor based in Belgrade, Serbia, and is associated with Constructor Technology AG. His patent work addresses machine-learning methods for helping autonomous vehicles perceive obstacles and drivable areas.

Latest Patents

Liubimov is credited with 1 patent, titled “Systems and Methods for Training a Camera-Based Perception Model Using Machine Learning.” The invention describes processing image and map data with a neural network to extract feature vectors. A transformer encoder converts those vectors from camera space into Bird’s Eye View (BEV) space. A detection head identifies objects, while a segmentation head produces a BEV map of objects and drivable surfaces. The system compares attributes from the two heads and updates the segmentation head’s weights, producing an updated BEV segmentation map.

Career Highlights

Liubimov’s listed patent reflects work on combining camera-based perception, map data, neural networks, and BEV representations for autonomous driving. He is associated with Constructor Technology AG.

Collaborations

His listed coworker is Ilya Shimchik. Their shared professional context is associated with Constructor Technology AG.

Conclusion

Maksim Liubimov’s patent focuses on training camera-based perception models to generate updated BEV maps that represent detected objects and drivable areas, contributing to research in autonomous-vehicle perception.