Changchang Wu



Home   |   Publications   |   Research   |   Code  | VisualSFM

I am a software engineer at Google, currently working on cardboard (VR). 3D computer vision has been my main research interest. 

I was previously a PhD student with Marc Pollefeys and Jan-Michael Frahm at UNC Chapel Hill. After graduation, I have worked as a postdoc with Steve Seitz and Brian Curless at University of Washington. 

NEW: Enliang Zheng, Changchang Wu, Structure from Motion using Structure-less Resection, ICCV 2015. (Is incremental SfM possible without 3-view overlap? Yes!)

VisualSFM : A Visual Structure From Motion System

A Visual 3D reconstruction system that integrates SiftGPU and Multicore Bundle Adjustment. It is easy to use with a feature-packed GUI. Check out my 3DV2013 paper for more algorithmic details:
"Towards Linear-time Incremental Structure from Motion", (webpage, supplemental, video)


Open-source Softwares for SfM

Multicore Bundle Adjustment, optimized for both GPU and CPU.

SIFT on GPU (SiftGPU), works for nVidia, ATI and Intel cards.