Guided Table Structure Recognition Through Anchor Optimization

This paper presents the novel approach towards table structure recognition by leveraging guided anchors. The concept differs from current state-of-the-art systems for that naively apply object detection methods. In contrast to prior techniques, first, we estimate viable anchors recognition. Subsequently, these are exploited locate rows and columns in tabular images. Furthermore, introduces a simple effective method improves results using layouts realistic scenarios. proposed is exhaustively evaluated on two publicly available datasets of recognition: ICDAR-2013 TabStructDB. Moreover, empirically established validity our implementing it previous approaches. We accomplished dataset with an average F1-measure 94.19% (92.06% 96.32% columns). Thus, relative error reduction more than 25% achieved. post-processing 95.46% 35%. surpassed baseline TabStructDB 94.57% (94.08% 95.06%