Beckman Institute Vision Research

  The Ponce Group

    Main Page
    Vision Research
    Robotics Research
    Technical Reports

Yasutaka Furukawa's Research

Carved Visual Hulls for High-Accuracy Image-Based Modeling
PowerPoint (39 MB, includes links to several videos), PostScript abstract
Presented as a technical sketch at SIGGRAPH 2005

See also: 3D Photography Dataset

From Textures to Objects

Learning Local Affine Representations for Texture and Object Recognition
PowerPoint (15 MB), PDF (10 MB)

Semi-Local Affine Parts for Object Recognition
PowerPoint (6 MB), PDF (3 MB)
Presented at the British Machine Vision Conference, September 2004, and the Snowbird Learning Workshop, April 2004 (Snowbird abstract)

Toward True 3D Object Recognition

PowerPoint (27 MB)
Presented at the Congrès de Reconnaissance des Formes et Intelligence Artificielle, Toulouse, France, January 2004

CVPR 2003 Presentations

3D Object Modeling and Recognition Using Affine-Invariant Patches and Multi-View Spatial Constraints
PDF (8 MB)

A Sparse Texture Representation Using Affine-Invariant Regions
PowerPoint (5 MB), PDF (5 MB)

Representing, Learning, and Recognizing Non-Rigid Textures and Texture Categories

PowerPoint (10 MB), PDF (2 MB)

Presented at the Snowbird Learning Workshop, April 2003

Fred Rothganger's Research

Projective Visual Hulls


Issues in 3D Object Recognition


Yakup Genc's Research -- Parametrized Image Varieties

Kitchen movie: input, synthesized
Flowers movie: input, synthesized
Box movie: input, synthesized
Face movie: input, synthesized
Tower movie: input, synthesized

Steve Sullivan's Research

Dinosaur modeling setup and fitted mesh (VRML)
Dinosaur movie (AVI)
Gargoyle modeling setup and fitted mesh (VRML)

Building Graphical Models of 3D Objects from Pictures


This work is supported in part by the UIUC Campus Research Board, the UIUC/CNRS Collaborative Research Agreement, and the National Science Foundation under grants IRI-990709, IIS-0308087, and IIS-0312438.