Technical Briefs

Modeling

Friday, 30 November 14:15 - 16:00 |  Garnet 219

3D Diff: An Interactive Approach to Mesh Differencing and Conflict Resolution - Picture

3D Diff: An Interactive Approach to Mesh Differencing and Conflict Resolution

We present an open source tool, 3D Diff, that supports differencing and merging of 3D models. We evaluate this tool with users and find the 3D Diff to be an effective way of merging 3D models.

Jozef Doboš, University College London
Anthony Steed, University College London


Adaptive Maximal Poisson-Disk Sampling on Surfaces - Picture

Adaptive Maximal Poisson-Disk Sampling on Surfaces

A novel approach for adaptive maximal Poisson-disk sampling on surfaces base on the power diagram and regular triangulation. The application for high quality remeshing.

Dong-Ming Yan, King Abdullah University of Science and Technology
Peter Wonka, King Abdullah University of Science and Technology


Dynamic 3-D Facial Compression Using Low Rank and Sparse Decomposition - Picture

Dynamic 3-D Facial Compression Using Low Rank and Sparse Decomposition

We propose a new compression framework for dynamic 3-D facial expressions acquired from structured light based 3-D camera. Low rank and sparse decomposition is applied to each dimension. An averaged 3-4 dB gain is achieved by the proposed scheme compared with existing algorithms.

Junhui Hou, Nanyang Technological University
Lap-Pui Chau, Nanyang Technological University
Ying He, Nanyang Technological University
Dao T. P. Quynh, Nanyang Technological University
Nadia Magnenat-Thalmann, Nanyang Technological University


Computing Defect-Insensitive Geodesic Distance on Broken Meshes - Picture

Computing Defect-Insensitive Geodesic Distance on Broken Meshes

This paper presents a new method to compute meaningful geodesic distance on broken meshes. The resulting distance is insensitive to the defects, such as shortcuts, holes, gaps, boundaries, mesh triangulation/resolution, and robust to noise.

Shi-Qing Xin, Ningbo University
Dao T.P. Quynh, Nanyang Technological University
Xiang Ying, Nanyang Technological University
Ying He, Nanyang Technological University


Basic Level Scene Understanding: From Labels to Structure and Beyond - Picture

Basic Level Scene Understanding: From Labels to Structure and Beyond

This paper summarizes our recent efforts toward visual scene understanding to unify recognition and reconstruction. We define a task called "basic level scene understanding" that is suitable for the current state of research in computer vision to produce a natural representation of the scene, and demonstrate some initial results for this task.

Jianxiong Xiao, Massachusetts Institute of Technology
Bryan C. Russell, University of Washington
James Hays, Brown University
Krista A. Ehinger, Massachusetts Institute of Technology
Aude Oliva, Massachusetts Institute of Technology
Antonio Torralba, Massachusetts Institute of Technology