Rahul Thomas
Papers
2
Total Citations
6
H-Index
2
About
Rahul Thomas is a rising researcher in computer vision and geometric deep learning, with a focused interest in the reverse engineering of 3D CAD models from visual data. His most significant contribution is the development of **Img2CAD**, a novel framework that leverages Vision-Language Models (VLMs) for conditional factorization, enabling the accurate reconstruction of parametric CAD models from single images. This work directly addresses the fundamental challenge of bridging the representational gap between unstructured 2D imagery and structured, editable CAD geometry. While his primary publication has garnered early citations, its impact is already evident in its potential to revolutionize workflows in interactive editing, manufacturing, architecture, and robotics. By integrating semantic understanding from VLMs, Thomas’s approach moves beyond traditional shape estimation, allowing for the extraction of meaningful design primitives and constraints. His work represents a critical step toward making 3D content creation more accessible and automated, positioning him as an innovator at the intersection of vision, language, and computational design.
Research Focus
Key Achievements
Top Papers
- 1
- 2