Daniel G. Aliaga

Assistant Professor of Computer Science

Purdue University

305 N. University St.

West Lafayette, IN 47907-2066

Office: (765) 496-7943

FAX: (765) 494-0739

Email: aliaga at cs purdue edu

Academic History

Research   

Pose-free Scene Reconstruction

Reconstructing large models from images is a significant challenge for computer graphics, computer vision, and related fields. In this project, we investigate an approach for simplifying the reconstruction process by mathematically eliminating external camera parameters. This results in less parameters to estimate and in an overall significantly more robust and accurate reconstruction. Unlike self-calibration, omitting pose parameters from the acquisition process implies no external calibration data must be computed or provided.
Modeling Repetitive Motions Creating models of dynamic 3D objects is an important part of content generation for computer graphics. If the states or poses of the dynamic object repeat often (but not necessarily periodically), we call such a repetitive motion. Our approach enables robustly acquiring and obtaining space-time 3D models of repetitive motions using as few as two cameras or one camera-projector pair.
Occlusion-Resistant Cameras Obtaining image sequences of popular and active environments is often hindered by unwanted interfering occluders. In this work, we propose a family of Occlusion-Resistant Camera (ORC) designs for acquiring such environments. Our cameras explicitly remove interfering occluders from acquired data in real-time and during live capture.

Rendering Urban Spaces

Our objective is to obtain models of large urban environments. We combine image-based capturing and rendering with procedural modeling techniques to allow the creation of novel structures in the style of real-world structures. We have developed techniques to capture building size environments and are pursuing the capture of larger urban spaces (e.g., urban layouts).

Sea of Images

 In this project, we present an image-based approach to providing interactive and photorealistic walkthroughs of complex indoor environments. Our strategy is to obtain a dense sampling of viewpoints in a large static environment with omnidirectional images and to replace the 3D reconstruction challenges with motorized-cart control, dense image-based sampling, and compression -- problems for which we can provide efficient solutions.

Graphics and Education

The project investigates novel and intuitive interfaces to help in educational scenarios. We are developing hardware and software tools for tabletop mixed-reality and for portable Tablet PCs in classrooms
Massive Model Rendering A key component of providing realism is rendering large and detailed 3D models at high frame rates. We explore various rendering acceleration methods, including visibility culling, geometry simplification, and image-based rendering.

[CGVLAB: Computer Graphics and Visualization Lab at Purdue University]

 

Publications

Conferences

Journal Articles

Dissertations/Theses

Software and Data

 

Courses

Number Title Year
 CS590G/CS635  Capturing, Modeling, Rendering 3D Structures  Fall 2003, Fall 2004, Spring 2007
 CS590M  Geometric Modeling and Applications  Spring 2004
 CS535  Interactive Computer Graphics  Fall 2005, Fall 2007
 CS490G  Tablet PC Graphics  Spring 2004, Spring 2005, Spring 2006
 CS397/CS497  Honor's Research  Fall 2004
 CS334  Fundamentals of Computer Graphics  Spring 2008
 CS251  Data Structures  Fall 2006

 

Professional Activities

Personal