Papers
28
Total Citations
1,039
H-Index
13
About
Federico Tombari is a prominent computer vision and robotics researcher whose work spans 3D scene understanding, neural representation learning, and embodied AI. Best known for his foundational contributions to neural fields — coordinate-based neural networks that parameterize physical scene properties across space and time — his 2022 survey on the topic has already garnered over 440 citations, reflecting the extraordinary momentum of this research direction. Tombari has made significant strides in bridging perception and action, from early work on affordance detection for robotic agents (2012) and simultaneous 3D reconstruction and object recognition in dense SLAM (2016), to more recent advances in 6-DoF robotic grasping from single RGB images and object rearrangement using scene graphs. His research also extends into multimodal AI, contributing to zero-shot reasoning frameworks that compose large pretrained models across vision and language. Work on dynamic object tracking, semantic mapping for autonomous vehicles, and NeRF-based shape and appearance reconstruction further demonstrates the breadth of his impact. With contributions cited hundreds of times across multiple subfields, Tombari stands as an influential figure shaping how machines perceive, reconstruct, and interact with the physical world.
Research Focus
Key Achievements
Top Papers
- 1Neural Fields in Visual Computing and Beyond447 citations · 2022
- 2Socratic Models: Composing Zero-Shot Multimodal Reasoning with Language171 citations · 2022
- 3
- 4MonoGraspNet: 6-DoF Grasping with a Single RGB Image42 citations · 2023
- 5
- 6
- 7
- 8
- 9Sensor substitution for video-based action recognition24 citations · 2016
- 10Lightweight Semantic Mesh Mapping for Autonomous Vehicles16 citations · 2021