No abstract

With the growing cosmopolitan culture of modern cities, the need of robust Multi-Lingual scene Text (MLT) detection and recognition systems has never been more immense.With the goal to systematically benchmark and push the stateof-the-art forward, the proposed competition builds on top of the RRC-MLT-2017 with an additional end-to-end task, an additional language in the real images dataset, a large scale multi-lingual synthetic dataset to assist the training, and a baseline End-to-End recognition method.The real dataset consists of 20,000 images containing text from 10 languages. The challenge has 4 tasks covering various aspects of multi-lingual scene text: (a) text detection, (b) cropped word script classification, (c) joint text detection and script classification and (d) end-to-end detection and recognition. In total, the competition received 60 submissions from the research and industrial communities. This paper presents the dataset, the tasks and the findings of the presented RRC-MLT-2019 challenge.

An end-to-end trainable (fully differentiable) method for multilanguage scene text localization and recognition is proposed. The approach is based on a single fully convolutional network (FCN) with shared layers for both tasks. E2E-MLT is the first published multi-language OCR for scene text. While trained in multi-language setup, E2E-MLT demonstrates competitive performance when compared to other methods trained for English scene text alone. The experiments show that obtaining accurate multi-language multi-script annotations is a challenging problem.

The analysis of protein-level multigene expression signature maps computed from the fusion of differently stained immunohistochemistry images is an emerging tool in cancer management. Creating these maps requires registering sets of histological images, a challenging task due to their large size, the non-linear distortions existing between consecutive sections and to the fact that the images correspond to different histological stains and thus, may have very different appearance. In this manuscript, we present a novel segmentation-based registration algorithm that exploits a multi-class pyramid and optimizes a fuzzy class assignment specially designed for this task. Compared to a standard nonrigid registration, the proposed method achieves an improved matching on both synthetic as well as real histological images of cancer lesions.

No abstract