J Zdenek, H Nakayama - Proceedings of the IEEE/CVF …, 2020 - openaccess.thecvf.com
… We use a separately trained scenetextdetector and an inpainting network. The scenetext detector predicts segmentation maps of text instances which are then used as masks for the …
… textdetection employing a semantic segmentation network but, since they cannot rely on a pixel–level supervision at the text … inside the bounding–box as text. Therefore, the scarcity of …
W Wu, J Xing, C Yang, Y Wang… - … Problems in Engineering, 2020 - Wiley Online Library
… by exploiting word annotations in rich large-scale real scenetext datasets. … weakly supervised semantic segmentation [34, 36–38], we propose a weakly supervisedscenetextdetector …
… networks for scenetext segmentation. … supervised learning approach is used to reduce the shift between training on real and synthetic data. Pixel–level supervisions for a textdetection …
C Xue, W Zhang, Y Hao, S Lu, PHS Torr… - European Conference on …, 2022 - Springer
… trained with weaksupervision only (ie partial texts in images without text bounding boxes). Our pre-trained model weights can be directly transferred to various scenetext detectors and …
E Metzenthin, C Bartz, C Meinel - arXiv preprint arXiv:2201.04866, 2022 - arxiv.org
… field of scenetextdetection requires complex … as weaksupervision, can reduce the amount of data needed. In this paper we propose a weaksupervision method for scenetextdetection, …
J Zhang, C Du, Z Feng, Y Wang… - … on Document Analysis …, 2019 - ieeexplore.ieee.org
… (1) The work is a further exploration of weaksupervision methods in scenetextdetection. (2) The network we designed uses a U-Net shape architecture rather than the CAM architecture…
K Ye, A Kovashka - … of the IEEE/CVF Conference on …, 2021 - openaccess.thecvf.com
… Our model uses the image’s paired caption as weaksupervision to … scene graphs without help from texts. To learn our model, we first allow context information to propagate on the text …
J Liu, Q Zhong, Y Yuan, H Su, B Du - Neurocomputing, 2020 - Elsevier
… is used for training textdetection model, has … supervisedscenetextdetection framework (SemiText), which trains robust and accurate scenetext detectors using a pre-trained supervised …