Share This Author
DualGAN: Unsupervised Dual Learning for Image-to-Image Translation
- Zili Yi, Hao Zhang, P. Tan, Minglun Gong
- Computer Science, BiologyIEEE International Conference on Computer Vision…
- 8 April 2017
A novel dual-GAN mechanism is developed, which enables image translators to be trained from two sets of unlabeled images from two domains, and can even achieve comparable or slightly better results than conditional GAN trained on fully labeled data.
Edge-aware point set resampling
- Hui Huang, Shihao Wu, Minglun Gong, D. Cohen-Or, U. Ascher, Hao Zhang
- Computer Science, MathematicsACM Trans. Graph.
- 7 February 2013
The Edge-Aware Resampling algorithm is demonstrated to be capable of producing consolidated point sets with noise-free normals and clean preservation of sharp features, and to lead to improved performance of edge-aware reconstruction methods and point set rendering techniques.
Stereoscopic inpainting: Joint color and depth completion from stereo images
- Liang Wang, Hailin Jin, Ruigang Yang, Minglun Gong
- Computer ScienceIEEE Conference on Computer Vision and Pattern…
- 23 June 2008
A novel algorithm takes stereo images and estimated disparity maps as input and fills in missing color and depth information introduced by occlusions or object removal and demonstrates the effectiveness of the proposed algorithm on several challenging data sets.
High-Quality Real-Time Stereo Using Adaptive Cost Aggregation and Dynamic Programming
- Liang Wang, Miao Liao, Minglun Gong, Ruigang Yang, D. Nistér
- Computer ScienceThird International Symposium on 3D Data…
- 1 June 2006
The key idea is simple: an adaptive aggregation step in a dynamic-programming (DP) stereo framework is introduced, which reduces the typical "streaking" artifacts without the penalty of blurry object boundaries.
A Performance Study on Different Cost Aggregation Approaches Used in Real-Time Stereo Matching
- Minglun Gong, Ruigang Yang, Liang Wang, Mingwei Gong
- Computer ScienceInternational Journal of Computer Vision
- 1 November 2007
Six recent cost aggregation approaches are implemented and optimized for graphics hardware so that real-time speed can be achieved and the performances of these aggregation approaches in terms of both processing speed and result quality are reported.
L1-medial skeleton of point cloud
A L1-medial skeleton construction algorithm is developed which can be directly applied to an unoriented raw point scan with significant noise, outliers, and large areas of missing data.
Near real-time reliable stereo matching using programmable graphics hardware
- Minglun Gong, Yee-Hong Yang
- Computer ScienceIEEE Computer Society Conference on Computer…
- 20 June 2005
A near-real-time stereo matching technique is presented in this paper, based on the reliability-based dynamic programming algorithm proposed earlier, which can generate semi-dense disparity maps using only two dynamic programming passes, while the previous approach requires 20-30 passes.
Quality-driven poisson-guided autoscanning
A quality-driven, Poisson-guided autonomous scanning method based on the analysis of a Poisson field and its geometric relation with an input scan to ensure the high quality scanning of the model.
Image-gradient-guided real-time stereo on graphics hardware
- Minglun Gong, Ruigang Yang
- Computer ScienceFifth International Conference on 3-D Digital…
- 13 June 2005
A real-time correlation-based stereo algorithm with improved accuracy that can run completely on the graphics board: from rectification, matching cost computation, cost aggregation, to the final disparity selection.
Real-Time Discriminative Background Subtraction
- Li Cheng, Minglun Gong, Dale Schuurmans, T. Caelli
- Computer ScienceIEEE Transactions on Image Processing
- 1 May 2011
The authors formulate background subtraction as minimizing a penalized instantaneous risk functional-yielding a local online discriminative algorithm that can quickly adapt to temporal changes and develop an implementation that can run efficiently on the highly parallel graphics processing unit (GPU).