Abstract: The extraction of relevant information from historical handwritten document collections is one of the key steps in order to make these manuscripts available for access and searches. In this competition, the goal is to detect the named entities and assign each of them a semantic category, and therefore, to simulate the filling in of a knowledge database. This paper describes the dataset, the tasks, the evaluation metrics, the participants methods and the results.

Abstract: Historical documents suffer from different types of degradation and noise such as background variation, uneven illumination or dark spots. In case of double-sided documents, another common problem is that the back side of the document usually interferes with the front side because of the transparency of the document or ink bleeding. This effect is called the show through phenomenon. Many methods are developed to solve these problems, and in the case of show-through, by scanning and matching both the front and back sides of the document. In contrast, our approach is designed to use only one side of the scanned document. We hypothesize that show-trough are low contrast components, while foreground components are high contrast ones. A Multiresolution Contrast (MC) decomposition is presented in order to estimate the contrast of features at different spatial scales. We cancel the show-through phenomenon by thresholding these low contrast components. This decomposition is also able to enhance the image removing shadowed areas by weighting spatial scales. Results show that the enhanced images improve the readability of the documents, allowing scholars both to recover unreadable words and to solve ambiguities.

Abstract: The adaptation capability to a wide range of domains is crucial for scene text spotting models when deployed to real-world conditions. However, existing state-of-the-art (SOTA) approaches usually incorporate scene text detection and recognition simply by pretraining on natural scene text datasets, which do not directly exploit the intermediate feature representations between multiple domains. Here, we investigate the problem of domain-adaptive scene text spotting, i.e., training a model on multi-domain source data such that it can directly adapt to target domains rather than being specialized for a specific domain or scenario. Further, we investigate a transformer baseline called Swin-TESTR to focus on solving scene-text spotting for both regular and arbitrary-shaped scene text along with an exhaustive evaluation. The results clearly demonstrate the potential of intermediate representations to achieve significant performance on text spotting benchmarks across multiple domains (e.g. language, synth-to-real, and documents). both in terms of accuracy and efficiency.

Abstract: The oral communication competence is defined on the top of the most relevant skills for one's professional and personal life. Because of the importance of communication in our activities of daily living, it is crucial to study methods to evaluate and provide the necessary feedback that can be used in order to improve these communication capabilities and, therefore, learn how to express ourselves better. In this work, we propose a system capable of evaluating quantitatively the quality of oral presentations in an automatic fashion. The system is based on a multi-modal RGB, depth, and audio data description and a fusion approach in order to recognize behavioral cues and train classifiers able to eventually predict communication quality levels. The performance of the proposed system is tested on a novel dataset containing Bachelor thesis' real defenses, presentations from an 8th semester Bachelor courses, and Master courses' presentations at Universitat de Barcelona. Using as groundtruth the marks assigned by actual instructors, our system achieves high performance categorizing and ranking presentations by their quality, and also making real-valued mark predictions.

Abstract: Although traditionally used in the machine translation field, the encoder-decoder framework has been recently applied for the generation of video and image descriptions. The combination of Convolutional and Recurrent Neural Networks in these models has proven to outperform the previous state of the art, obtaining more accurate video descriptions. In this work we propose pushing further this model by introducing two contributions into the encoding stage. First, producing richer image representations by combining object and location information from Convolutional Neural Networks and second, introducing Bidirectional Recurrent Neural Networks for capturing both forward and backward temporal relationships in the input frames.