Research
Spatial Intelligence
Developing models that allows digitization and spatial understanding. We develop Computer Vision models that grant semantic understanding of 3D space. understanding geometry, physics, object affordances, and spatial relationships.
Perception & Navigation
By bridging the gap between raw sensory input and spatial reasoning, we are building the foundational models for next-generation Autonomous Agents and context-aware Augmented Reality. This allows systems to navigate, interact with, and reason about the physical world with human-like intelligence.
Generative World Models
We are pioneering in the research & develop of LWMS (Large World Models) capable of synthesizing interactive 3D environments from textual or sparse visual inputs.
Unlike traditional generative video, our models learn the underlying physics and causal consistency of reality. This allows us to simulate potential futures and generate coherent, explorable worlds, revolutionizing simulation training and AI understanding.
World digital representation
We are advancing the state of the art in 3D reconstruction. Moving beyond traditional photogrammetry, our research focuses on 3D Gaussian Splatting and Implicit Neural Representations to capture the physical world with perfect accuracy.
Models designed to digitize complex environments from cultural heritage sites to large-scale urban areas-preserving high-frequency details and material properties to enable photo-realistic, real-time rendering on any device.
