method: DMText_lsvt2019-04-30
Authors: Pei Xu, Shan Huang. (Tencent)
Description: This is an instance segmentation based method. We segment the text polygons from region proposals. In the training process, We first train on images synthesized by the method of VGG-synthtext, and then fine tune on the LSVT training images. In the post-processing, we adopt NMS and filter out those polygons with low scores.
method: Fudan-Supremind Detection2019-04-30
Authors: Yaowu Wei, Shangchao Su, Tairu Qiu, Xunyan Wang, Shaokang Lin, Zili Yi, Lei Deng, Mulin Xu, Jianqi Ma, Bin Li, Xiangyang Xue
Description: We propose a text segmentation algorithm based on Cascaded Mask-RCNN with deformable convolution and DenseASPP. And furthermore, we do some changes.
method: SRCB_LSVT2019-04-29
Authors: Yi Yu, Haiyang Guo, Xiaobing Wang, Yingying Jiang
Description: We use Mask R-CNN for text detection in our submission, which extends Faster R-CNN by adding a branch for predicting an object mask in parallel with the existing branch for bounding box recognition. Mask R-CNN is simple to train and adds only a small overhead to Faster R-CNN. And Mask R-CNN is the state-of-the-art object detection framework now. Therefore, it is used here for text detection task. We use the Mask R-CNN in https://github.com/facebookresearch/Detectron and the backbone network is ResNext 101. Meanwhile, the object is text here and the number of classification classes is 2. Besides, polygon based NMS is used for post-processing to remove overlapped text regions.
Date | Method | Recall | Precision | Hmean | |||
---|---|---|---|---|---|---|---|
2019-04-30 | DMText_lsvt | 80.03% | 89.28% | 84.40% | |||
2019-04-30 | Fudan-Supremind Detection | 81.81% | 85.94% | 83.82% | |||
2019-04-29 | SRCB_LSVT | 77.95% | 89.48% | 83.32% | |||
2020-11-23 | hrnet_w40_casecade | 72.88% | 73.51% | 73.19% | |||
2019-04-29 | test4 | 61.75% | 80.66% | 69.95% | |||
2019-04-25 | AdvancedEast model with post processing | 54.75% | 66.55% | 60.08% |