Current and past offered theses

Current projects (Academic year 2026 - 2027)

Space domain

• Binary Neural Networks for Post-Disaster Change Detection on FPGA
• Deep Learning for Telemetry Data Analysis in Satellite Missions
• Edge AI for Space Debris Streaks Detection
• Efficient Multiscale Onboard Vision for Rapid Building Damage Assessment after Disasters
• Hardware-Aware Neural Architecture Search for Satellite Telemetry Anomaly Detection
• Towards FPGA-Deployable Continual Learning for Satellite Telemetry Anomaly Detection

CERN / Particle physics domain

• Hardware acceleration of deep learning models for particle tracking at the LHC and beyond
• Model compression and optimization of deep learning architectures for particle tracking at the LHC and beyond

Foundational research

• Optimized hybrid CNN-Transformer architectures for Pareto-efficient vision systems
• Exploring logarithmic quantization for efficient neural network inference
• Hardware-Aware Low-Rank Factorization
• One shot vs. iterative pruning
• Study and comparative evaluation of vector and accelerator implementations for the RISC-V architecture

Past projects (Academic years 2024 - 2026)

• BeeAIve: Machine Learning for Queen Presence Classification in Beehive Audio with Edge Deployment
• Edge AI on Embedded Systems for Gray Mold Disease Detection
• Collaborative weed detection at the edge using federated learning
• Edge AI on Embedded Systems for Pest Detection in Tomato Plants