A Digitalized Recomposition Technique Based on Photo Quality Evaluation Criteria

摘要

Photographic evaluation is based mainly on qualitative factors which are very personal and subjective. The qualitative factors must be converted into quantitative measures, however, in order to develop a digitalized recomposition algorithm. In addition, the theories of composition in photography should be explored for digitizing photo quality evaluation criteria. In this paper, a new evaluation algorithm for photographic recomposition based on photo quality evaluation principles will be presented. In detail, we formulate the rule-of-thirds as an optimization problem of the feature vector in the image. Simplicity factors can be formulated as a calculation problem of the size of region of interest (ROI) segments. The size of ROI and the moving direction of foreground object are used to formulate the rule of space. The presented algorithm can be extended to photographic evaluation field. Algorithmic excellence will be proven by experimental results. The proposed technique is fully automatic while previous works need manual interaction. The authors expect that the presented algorithm can be applied in the near future since many related state-of-the-art technologies are embedded on commercial cameras these days.