|
Ruben Tito, Dimosthenis Karatzas and Ernest Valveny. 2023. Hierarchical multimodal transformers for Multi-Page DocVQA. PR, 144, 109834.
Abstract: Document Visual Question Answering (DocVQA) refers to the task of answering questions from document images. Existing work on DocVQA only considers single-page documents. However, in real scenarios documents are mostly composed of multiple pages that should be processed altogether. In this work we extend DocVQA to the multi-page scenario. For that, we first create a new dataset, MP-DocVQA, where questions are posed over multi-page documents instead of single pages. Second, we propose a new hierarchical method, Hi-VT5, based on the T5 architecture, that overcomes the limitations of current methods to process long multi-page documents. The proposed method is based on a hierarchical transformer architecture where the encoder summarizes the most relevant information of every page and then, the decoder takes this summarized information to generate the final answer. Through extensive experimentation, we demonstrate that our method is able, in a single stage, to answer the questions and provide the page that contains the relevant information to find the answer, which can be used as a kind of explainability measure.
|
|
|
Pau Riba, Josep Llados and Alicia Fornes. 2020. Hierarchical graphs for coarse-to-fine error tolerant matching. PRL, 134, 116–124.
Abstract: During the last years, graph-based representations are experiencing a growing usage in visual recognition and retrieval due to their ability to capture both structural and appearance-based information. Thus, they provide a greater representational power than classical statistical frameworks. However, graph-based representations leads to high computational complexities usually dealt by graph embeddings or approximated matching techniques. Despite their representational power, they are very sensitive to noise and small variations of the input image. With the aim to cope with the time complexity and the variability present in the generated graphs, in this paper we propose to construct a novel hierarchical graph representation. Graph clustering techniques adapted from social media analysis have been used in order to contract a graph at different abstraction levels while keeping information about the topology. Abstract nodes attributes summarise information about the contracted graph partition. For the proposed representations, a coarse-to-fine matching technique is defined. Hence, small graphs are used as a filtering before more accurate matching methods are applied. This approach has been validated in real scenarios such as classification of colour images or retrieval of handwritten words (i.e. word spotting).
Keywords: Hierarchical graph representation; Coarse-to-fine graph matching; Graph-based retrieval
|
|
|
Klaus Broelemann, Anjan Dutta, Xiaoyi Jiang and Josep Llados. 2012. Hierarchical graph representation for symbol spotting in graphical document images. Structural, Syntactic, and Statistical Pattern Recognition, Joint IAPR International Workshop. Springer Berlin Heidelberg, 529–538. (LNCS.)
Abstract: Symbol spotting can be defined as locating given query symbol in a large collection of graphical documents. In this paper we present a hierarchical graph representation for symbols. This representation allows graph matching methods to deal with low-level vectorization errors and, thus, to perform a robust symbol spotting. To show the potential of this approach, we conduct an experiment with the SESYD dataset.
|
|
|
Nuria Cirera, Alicia Fornes and Josep Llados. 2015. Hidden Markov model topology optimization for handwriting recognition. 13th International Conference on Document Analysis and Recognition ICDAR2015.626–630.
Abstract: In this paper we present a method to optimize the topology of linear left-to-right hidden Markov models. These models are very popular for sequential signals modeling on tasks such as handwriting recognition. Many topology definition methods select the number of states for a character model based
on character length. This can be a drawback when characters are shorter than the minimum allowed by the model, since they can not be properly trained nor recognized. The proposed method optimizes the number of states per model by automatically including convenient skip-state transitions and therefore it avoids the aforementioned problem.We discuss and compare our method with other character length-based methods such the Fixed, Bakis and Quantile methods. Our proposal performs well on off-line handwriting recognition task.
|
|
|
Alloy Das, Sanket Biswas, Ayan Banerjee, Josep Llados, Umapada Pal and Saumik Bhattacharya. 2024. Harnessing the Power of Multi-Lingual Datasets for Pre-training: Towards Enhancing Text Spotting Performance. Winter Conference on Applications of Computer Vision.718–728.
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.
|
|
|
Jon Almazan, Albert Gordo, Alicia Fornes and Ernest Valveny. 2013. Handwritten Word Spotting with Corrected Attributes. 15th IEEE International Conference on Computer Vision.1017–1024.
Abstract: We propose an approach to multi-writer word spotting, where the goal is to find a query word in a dataset comprised of document images. We propose an attributes-based approach that leads to a low-dimensional, fixed-length representation of the word images that is fast to compute and, especially, fast to compare. This approach naturally leads to an unified representation of word images and strings, which seamlessly allows one to indistinctly perform query-by-example, where the query is an image, and query-by-string, where the query is a string. We also propose a calibration scheme to correct the attributes scores based on Canonical Correlation Analysis that greatly improves the results on a challenging dataset. We test our approach on two public datasets showing state-of-the-art results.
|
|
|
David Fernandez. 2010. Handwritten Word Spotting in Old Manuscript Images using Shape Descriptors. (Master's thesis, .)
|
|
|
David Fernandez, Josep Llados and Alicia Fornes. 2011. Handwritten Word Spotting in Old Manuscript Images Using a Pseudo-Structural Descriptor Organized in a Hash Structure. In Jordi Vitria, Joao Miguel Raposo and Mario Hernandez, eds. 5th Iberian Conference on Pattern Recognition and Image Analysis.628–635.
Abstract: There are lots of historical handwritten documents with information that can be used for several studies and projects. The Document Image Analysis and Recognition community is interested in preserving these documents and extracting all the valuable information from them. Handwritten word-spotting is the pattern classification task which consists in detecting handwriting word images. In this work, we have used a query-by-example formalism: we have matched an input image with one or multiple images from handwritten documents to determine the distance that might indicate a correspondence. We have developed an approach based in characteristic Loci Features stored in a hash structure. Document images of the marriage licences of the Cathedral of Barcelona are used as the benchmarking database.
|
|
|
Pau Riba, Josep Llados and Alicia Fornes. 2015. Handwritten Word Spotting by Inexact Matching of Grapheme Graphs. 13th International Conference on Document Analysis and Recognition ICDAR2015.781–785.
Abstract: This paper presents a graph-based word spotting for handwritten documents. Contrary to most word spotting techniques, which use statistical representations, we propose a structural representation suitable to be robust to the inherent deformations of handwriting. Attributed graphs are constructed using a part-based approach. Graphemes extracted from shape convexities are used as stable units of handwriting, and are associated to graph nodes. Then, spatial relations between them determine graph edges. Spotting is defined in terms of an error-tolerant graph matching using bipartite-graph matching algorithm. To make the method usable in large datasets, a graph indexing approach that makes use of binary embeddings is used as preprocessing. Historical documents are used as experimental framework. The approach is comparable to statistical ones in terms of time and memory requirements, especially when dealing with large document collections.
|
|
|
Juan Ignacio Toledo, Sebastian Sudholt, Alicia Fornes, Jordi Cucurull, A. Fink and Josep Llados. 2016. Handwritten Word Image Categorization with Convolutional Neural Networks and Spatial Pyramid Pooling. Joint IAPR International Workshops on Statistical Techniques in Pattern Recognition (SPR) and Structural and Syntactic Pattern Recognition (SSPR). Springer International Publishing, 543–552. (LNCS.)
Abstract: The extraction of relevant information from historical document collections is one of the key steps in order to make these documents available for access and searches. The usual approach combines transcription and grammars in order to extract semantically meaningful entities. In this paper, we describe a new method to obtain word categories directly from non-preprocessed handwritten word images. The method can be used to directly extract information, being an alternative to the transcription. Thus it can be used as a first step in any kind of syntactical analysis. The approach is based on Convolutional Neural Networks with a Spatial Pyramid Pooling layer to deal with the different shapes of the input images. We performed the experiments on a historical marriage record dataset, obtaining promising results.
Keywords: Document image analysis; Word image categorization; Convolutional neural networks; Named entity detection
|
|