Leandro Di Bella

3rd-year PhD student · ETRO, Vrije Universiteit Brussel (VUB)

Autonomous perception

Computer vision & generative AI for driving & drones.

I work on deployable perception systems with strong temporal consistency and practical robustness.

Multi-object tracking: stable 3D MOT for autonomy.
Vision-language: grounding & reasoning for perception.
Forecasting: generative motion prediction in complex scenes.

CV Publications Projects

HybridTrack: A Hybrid Approach for Robust Multi-Object Tracking IEEE Robotics and Automation Letters (RA-L/ICRA)

ReferGPT: Towards Zero-Shot Referring Multi-Object Tracking IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)

DeepKalPose: An Enhanced Deep-Learning Kalman Filter for Temporally Consistent Monocular Vehicle Pose Estimation Electronics Letters

ChronoFusion: Spatio-Temporal Super-Resolution based on Graph VAEs and Gated Fusion European Workshop on Visual Information Processing (EUVIP)

What I work on

Autonomous perception Multi-object tracking Vision-language models Generative forecasting Autonomous driving Autonomous drones Sensor fusion

Current focus: real-time robustness, temporal consistency, and reliable behavior under distribution shift.

Mappx

Map-centric social travel app (1000+ users).

Flutter FastAPI PostgreSQL Azure Mapbox

Open to full-time roles in Belgium’s research & engineering ecosystem.

leandro.dibella@gmail.com GitHub LinkedIn Scholar