3D understanding

Modeling the Visual World Across Space and Time: From Observation to Interaction

-
This dissertation develops methods to recover and model the visual world consistently across space and time from images and video, enabling both accurate understanding of scene geometry and coherent, event-driven video editing grounded in user intent.

Geometry-Aware Visual Understanding and Stereoscopic Video Generation

-

B1 R2202

This thesis develops a geometry-aware dissolving transformation that selectively removes unnecessary fine-grained visual and geometric detail while preserving coarse structural information, to achieve more robust representation learning, higher-quality stereoscopic video synthesis, and more efficient multi-view geometry processing.

Jian Shi

Ph.D. Student, Computer Science
Jian Shi is a final-year Ph.D. candidate in Computer Science at King Abdullah University of Science and Technology (KAUST), where he is advised by Prof. Peter Wonka, specializing in computer vision, geometric deep learning, generative AI, and 3D visual computing, with particular emphasis on stereo video synthesis, 3D perception, scene reconstruction, and immersive visual experiences.

Silvio Giancola

Silvio Giancola is a Research Scientist at King Abdullah University of Science and Technology (KAUST), under the supervision of Prof. Bernard Ghanem in the Image and Video Understanding Laboratory (IVUL), part of the Center of Excellence for Generative AI (GenAI).