About Me

I’m Ke Wang (王珂), a PhD student in the 3D Computer Vision Group, Department of Computer Science, UNC Chapel Hill. I’m advised by Dr. Jan-Michael Frahm and Dr. Enrique Dunn.

My research focuses on 3D computer vision, including depth map computations, 3D reconstruction from satellite images, large-scale video data association, image geo-localization, etc. Recently I’ve been using deep neural networks to perform parametric/semantic reconstructions from aerial/satellite imagery.


Education


Publications

  1. Ke Wang, Mohit Bansal, and Jan-Michael Frahm: Retweet wars: Tweet popularity prediction via dynamic multimodal regression. IEEE Winter Conference on Applications of Computer Vision (WACV), 2018.
  2. Ke Wang, Enrique Dunn, Mikel Rodriguez, and Jan-Michael Frahm: Efficient video collection association using geometry-aware bag-of-iconics representations. IPSJ Transactions on Computer Vision and Applications, 2017.
  3. Ke Wang and Jan-Michael Frahm: Fast and accurate satellite multi-view stereo using edge-aware interpolation. International Conference on 3D Vision (3DV), 2017.
  4. Ke Wang and Jan-Michael Frahm: Single image parametric building model estimation from satellite imagery. International Conference on 3D Vision (3DV), 2017.
  5. Ke Wang, Enrique Dunn, Mikel Rodriguez, and Jan-Michael Frahm: Bringing 3d models together: Mining video liaisons in crowdsourced reconstructions. The 13th Asian Conference on Computer Vision (ACCV), 2016. [Paper] [Bibtex] [Poster] [Video]
  6. Ke Wang, Craig Stutts, Enrique Dunn, and Jan-Michael Frahm: Efficient joint stereo estimation and land usage classification for multiview satellite data. IEEE Winter Conference on Applications of Computer Vision (WACV), 2016. [Paper] [Bibtex]
  7. Enliang Zheng, Ke Wang, Enrique Dunn, and Jan-Michael Frahm: Minimal solvers for 3d geometry from satellite imagery. IEEE International Conference on Computer Vision (ICCV), 2015. [Paper] [Bibtex]
  8. Enliang Zheng, Ke Wang, Enrique Dunn, and Jan-Michael Frahm: Joint object class sequencing and trajectory triangulation (jost). European Conference on Computer Vision (ECCV) (Oral), 2014. [Paper] [Bibtex]
  9. Yilin Wang, Ke Wang, Enrique Dunn, and Jan-Michael Frahm: Stereo under sequential optimal sampling: A statistical analysis framework for search space reduction. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2014. [Paper] [Bibtex] [Video]
  10. Ke Wang, Enrique Dunn, Joseph Tighe, and Jan-Michael Frahm: Combining semantic scene priors and haze removal for single image depth estimation. IEEE Winter Conference on Applications of Computer Vision (WACV), 2014. [Paper] [Bibtex]

Industry Experience