Active Document Layout Synthesis

Lin, Xiaofan

HP Development Company

关键词: variable data printing;    constraint solving;    document layout synthesis;    table formatting;    Simplex;   

RP-ID  :  HPL-2005-106

学科分类：计算机科学（综合）

美国|英语

来源: HP Labs

【 摘 要 】

Document layout analysis has been researched for many years. However, there is little work on the reverse of document layout analysis: document layout synthesis, whose goal is to generate logically correct and aesthetically appealing layout given the text/image contents and the flexible layout template. This paper introduces a new automatic document layout synthesis method, which can actively pursue the optimal text block height-width tradeoff simultaneously with the block position adjustment. The key idea is to use a cluster of linear functions to approximate the nonlinear height-width curve. Then two-pass Simplex algorithm is used to solve the layout synthesis problem. This method has awide range of applications, such as Variable Data Printing (VDP), automatic table formatting, and automating document layout ground truth generation. Notes: To be published in and presented at the 8th International Conference on Document Analysis and Recognition, 29 August - 1 September 2005, Seoul, Korea 5 Pages