Prof. C V Jawahar and his students Madhav Agarwal and Ajoy Mondal presented a paper virtually on CDeC-Net: Composite Deformable Cascade Network for Table Detection in Document Images at the 25th International Conference on Pattern Recognition (ICPR-2020) in Milano, Italy from 10 – 15 January. Prof. C V Jawahar and his students Madhav Agarwal and Ajoy Mondal explain their research work:
Localizing page elements/objects such as tables, figures, equations, etc. is the primary step in extracting information from document images. We propose a novel end-to-end trainable deep network, (CDeC-Net) for detecting tables present in the documents. The proposed network consists of a multistage extension of Mask R-CNN with a dual backbone having deformable convolution for detecting tables varying in scale with high detection accuracy at higher IoU threshold. We empirically evaluate CDeC-Net on all the publicly available benchmark datasets – ICDAR-2013, ICDAR-2017, ICDAR-2019,UNLV, Marmot, PubLayNet, and TableBank – with extensive experiments.
Our solution has three important properties: (i) a single trained model CDeC-Net‡ performs well across all the popular benchmark datasets; (ii) we report excellent performances across multiple, including higher, thresholds of IoU; (iii) by following the same protocol of the recent papers for each of the benchmarks, we consistently demonstrate the superior quantitative performance. Our code and models will be publicly released for enabling the reproducibility of the results.
ICPR2020 is the flagship conference of IAPR the International Association of Pattern Recognition and the premiere conference in pattern recognition, covering computer vision, image, sound, speech, sensor patterns processing and machine intelligence. It was a 6 day event that comprised the main conference, tutorials and co-located workshops. ICPR2020 was the 25th event of the series and turned 50 years old since its beginning. It provided a great opportunity to nurture new ideas and collaborations for students and researchers.