My research interests lie at the intersection of Computer Vision, Deep Learning, and 3D Reconstruction, with a particular focus on VFMs, and Generative Models.
During my MSc (2023–2025), I was advised by Sagie Benaim, researching methods to instill 3D awareness into 2D Vision Foundation Models (VFMs) via Gaussian Splatting.
Previously, I served as an ML Research Assistant at Sheba Medical Center (2021–2023), developing AI-based diagnostic tools for breast MRI. I specialized in applying deep learning for high-sensitivity tumor classification and breast tissue segmentation using PyTorch.
Splat and Distill instills 3D awareness into 2D VFMs by lifting features into 3D Gaussians and splatting views, boosting geometric and semantic representation.
Introducing a method for distilling structurally disentangled feature fields, enabling more precise 3D scene understanding and editing capabilities
Using AI-based analysis on temporal MRI sequences to identify high-risk signs of breast cancer in mutation carriers earlier than traditional methods
Feel free to reach out:
davidshavin4 at gmail dot com