Abstract: 0,624 JCR
Word spotting is the process of retrieving all instances of a queried keyword from a digital library of document images. In this paper we evaluate the performance of different word descriptors to assess the advantages and disadvantages of statistical and structural models in a framework of query-by-example word spotting in historical documents. We compare four word representation models, namely sequence alignment using DTW as a baseline reference, a bag of visual words approach as statistical model, a pseudo-structural model based on a Loci features representation, and a structural approach where words are represented by graphs. The four approaches have been tested with two collections of historical data: the George Washington database and the marriage records from the Barcelona Cathedral. We experimentally demonstrate that statistical representations generally give a better performance, however it cannot be neglected that large descriptors are difficult to be implemented in a retrieval scenario where word spotting requires the indexation of data with million word images.

Abstract: We present a method for spotting a subgraph in a graph repository. Subgraph spotting is a very interesting research problem for various application domains where the use of a relational data structure is mandatory. Our proposed method accomplishes subgraph spotting through graph embedding. We achieve automatic indexation of a graph repository during off-line learning phase, where we (i) break the graphs into 2-node sub graphs (a.k.a. cliques of order 2), which are primitive building-blocks of a graph, (ii) embed the 2-node sub graphs into feature vectors by employing our recently proposed explicit graph embedding technique, (iii) cluster the feature vectors in classes by employing a classic agglomerative clustering technique, (iv) build an index for the graph repository and (v) learn a Bayesian network classifier. The subgraph spotting is achieved during the on-line querying phase, where we (i) break the query graph into 2-node sub graphs, (ii) embed them into feature vectors, (iii) employ the Bayesian network classifier for classifying the query 2-node sub graphs and (iv) retrieve the respective graphs by looking-up in the index of the graph repository. The graphs containing all query 2-node sub graphs form the set of result graphs for the query. Finally, we employ the adjacency matrix of each result graph along with a score function, for spotting the query graph in it. The proposed subgraph spotting method is equally applicable to a wide range of domains, offering ease of query by example (QBE) and granularity of focused retrieval. Experimental results are presented for graphs generated from two repositories of electronic and architectural document images.

Abstract: Structural pattern recognition approaches offer the most expressive, convenient, powerful but computational expensive representations of underlying relational information. To benefit from mature, less expensive and efficient state-of-the-art machine learning models of statistical pattern recognition they must be mapped to a low-dimensional vector space. Our method of explicit graph embedding bridges the gap between structural and statistical pattern recognition. We extract the topological, structural and attribute information from a graph and encode numeric details by fuzzy histograms and symbolic details by crisp histograms. The histograms are concatenated to achieve a simple and straightforward embedding of graph into a low-dimensional numeric feature vector. Experimentation on standard public graph datasets shows that our method outperforms the state-of-the-art methods of graph embedding for richly attributed graphs.

Abstract: We call on the Document AI (DocAI) community to re-evaluate current methodologies and embrace the challenge of creating more practically-oriented benchmarks. Document Understanding Dataset and Evaluation (DUDE) seeks to remediate the halted research progress in understanding visually-rich documents (VRDs). We present a new dataset with novelties related to types of questions, answers, and document layouts based on multi-industry, multi-domain, and multi-page VRDs of various origins and dates. Moreover, we are pushing the boundaries of current methods by creating multi-task and multi-domain evaluation setups that more accurately simulate real-world situations where powerful generalization and adaptation under low-resource settings are desired. DUDE aims to set a new standard as a more practical, long-standing benchmark for the community, and we hope that it will lead to future extensions and contributions that address real-world challenges. Finally, our work illustrates the importance of finding more efficient ways to model language, images, and layout in DocAI.

Abstract: Until recently, most computer vision models remained illiterate, largely ignoring the semantically rich and explicit information contained in scene text. Recent progress in scene text detection and recognition has recently allowed exploring its role in a diverse set of open computer vision problems, e.g. image classification, image-text retrieval, image captioning, and visual question answering to name a few. The explicit semantics of scene text closely requires specific modeling similar to language. However, scene text is a particular signal that has to be interpreted according to a comprehensive perspective that encapsulates all the visual cues in an image. Incorporating this information is a straightforward task for humans, but if we are unfamiliar with a language or scripture, achieving a complete world understanding is impossible (e.a. visiting a foreign country with a different alphabet). Despite the importance of scene text, modeling it requires considering the several ways in which scene text interacts with an image, processing and fusing an additional modality. In this thesis, we mainly focus
on two tasks, scene text-based fine-grained image classification, and cross-modal retrieval. In both studied tasks we identify existing limitations in current approaches and propose plausible solutions. Concretely, in each chapter: i) We define a compact way to embed scene text that generalizes to unseen words at training time while performing in real-time. ii) We incorporate the previously learned scene text embedding to create an image-level descriptor that overcomes optical character recognition (OCR) errors which is well-suited to the fine-grained image classification task. iii) We design a region-level reasoning network that learns the interaction through semantics among salient visual regions and scene text instances. iv) We employ scene text information in image-text matching and introduce the Scene Text Aware Cross-Modal retrieval StacMR task. We gather a dataset that incorporates scene text and design a model suited for the newly studied modality. v) We identify the drawbacks of current retrieval metrics in cross-modal retrieval. An image captioning metric is proposed as a way of better evaluating semantics in retrieved results. Ample experimentation shows that incorporating such semantics into a model yields better semantic results while
requiring significantly less data to converge.