Lucas-Kanade 20 Years On: A Unifying Framework
International Journal of Computer Vision
International Journal on Document Analysis and Recognition
Content-preserving warps for 3D video stabilization
ACM SIGGRAPH 2009 papers
Manuscript Bleed-through Removal via Hysteresis Thresholding
ICDAR '09 Proceedings of the 2009 10th International Conference on Document Analysis and Recognition
Recto-verso Registration, Enhancement and Segmentation of Ancient Documents
VSMM '09 Proceedings of the 2009 15th International Conference on Virtual Systems and Multimedia
IEEE Transactions on Pattern Analysis and Machine Intelligence
A Variational Approach to Degraded Document Enhancement
IEEE Transactions on Pattern Analysis and Machine Intelligence
User-assisted ink-bleed reduction
IEEE Transactions on Image Processing - Special section on distributed camera networks: sensing, processing, communication, and implementation
Non-rigid Registration and Restoration of Double-Sided Historical Manuscripts
ICDAR '11 Proceedings of the 2011 International Conference on Document Analysis and Recognition
Document Recto-verso Registration Using a Dynamic Time Warping Algorithm
ICDAR '11 Proceedings of the 2011 International Conference on Document Analysis and Recognition
A Non-parametric Framework for Document Bleed-through Removal
CVPR '13 Proceedings of the 2013 IEEE Conference on Computer Vision and Pattern Recognition
Hi-index | 0.00 |
Accurate registration of document recto and verso sides with bleed-through degradation is essential for accurate automatic non-blind bleed-through removal. This paper presents a registration method for documents with bleed-through degradation, and also an objective registration evaluation scheme. In the proposed method the two sides are first globally aligned using the outline of the page, and then a local grid point warp is applied using the sum of squared difference between both the image intensity and gradient fields as an error metric, and a content preserving smoothness penalty. The displacement fields of the combined global and local warps are then evaluated against manually registered full document image displacement fields and compared with recent document registration methods.