Research
ELLIS fosters international collaboration across domains, connecting top researchers while investing in the next generation of AI talent.
Members
ELLIS Members are leading scientists in machine learning and AI, shaping Europe's global position in these fields.
About
ELLIS is a network of excellence connecting top AI researchers across European borders to strengthen the leadership of AI made in Europe.
Ido Sobol
PhD
Technion - Israel Institute of Technology (Technion)
Controllable and Realistic 3D Generation through Cross-Domain Training
Real-world data provides unmatched realism but rarely comes with the paired 3D control signals (such as normals, depth, or material properties) needed for conditioning diffusion models. Synthetic 3D assets, on the other hand, offer perfect access to such signals but suffer from a lack of photorealism. This project aims to bridge these two worlds by developing training methods that combine realism from real data with controllability from synthetic data, enabling diffusion models that are both photorealistic and precisely controllable.
Track:
Academic Track
Exchange Advisor
PhD Duration:
October 1st, 2025 - October 1st, 2028
Research
About
