Image retrieval can be formulated as a ranking problem where the goal is to order database images by decreasing similarity to the query. Recent deep models for image retrieval have outperformed traditional methods by leveraging ranking-tailored loss functions, but important theoretical and practical problems remain. First, rather than directly optimizing the global ranking, they minimize an upper-bound on the essential loss, which does not necessarily result in an optimal mean average precision (mAP). Second, these methods require significant engineering efforts to work well, e.g., special pre-training and hard-negative mining. In this paper we propose instead to directly optimize the global mAP by leveraging recent advances in listwise loss formulations. Using a histogram binning approximation, the AP can be differentiated and thus employed to end-to-end learning. Compared to existing losses, the proposed method considers thousands of images simultaneously at each iteration and eliminates the need for ad hoc tricks. It also establishes a new state of the art on many standard retrieval benchmarks. Models and evaluation scripts have been made available at: https://europe.naverlabs.com/Deep-Image-Retrieval/.

The goal of this work is to bring semantics into the tasks of text recognition and retrieval in natural images. Although text recognition and retrieval have received a lot of attention in recent years, previous works have focused on recognizing or retrieving exactly the same word used as a query, without taking the semantics into consideration. In this paper, we ask the following question: can we predict semantic concepts directly from a word image, without explicitly trying to transcribe the word image or its characters at any point? For this goal we propose a convolutional neural network (CNN) with a weighted ranking loss objective that ensures that the concepts relevant to the query image are ranked ahead of those that are not relevant. This can also be interpreted as learning a Euclidean space where word images and concepts are jointly embedded. This model is learned in an end-to-end manner, from image pixels to semantic concepts, using a dataset of synthetically generated word images and concepts mined from a lexical database (WordNet). Our results show that, despite the complexity of the task, word images and concepts can indeed be associated with a high degree of accuracy.

This paper presents a prototype system running on portable devices for browsing and word searching through historical handwritten document collections. The platform adapts the paradigm of eBook reading, where the narrative is not necessarily sequential, but centered on the user actions. The novelty is to replace digitally born books by digitized historical manuscripts of marriage licenses, so document analysis tasks are required in the browser. With an active reading paradigm, the user can cast queries of people names, so he/she can implicitly follow genealogical links. In addition, the system allows combined searches: the user can refine a search by adding more words to search. As a second contribution, the retrieval functionality involves as a core technology a word spotting module with an unified approach, which allows combined query searches, and also two input modalities: query-by-example, and query-by-string.

This paper presents an image database of historical handwritten marriages records stored in the archives of Barcelona cathedral, and the corresponding meta-data addressed to evaluate the performance of document analysis algorithms. The contribution of this paper is twofold. First, it presents a complete ground truth which covers the whole pipeline of handwriting recognition research, from layout analysis to recognition and understanding. Second, it is the first dataset in the emerging area of genealogical document analysis, where documents are manuscripts pseudo-structured with specific lexicons and the interest is beyond pure transcriptions but context dependent.

This paper addresses the problems of word spotting and word recognition on images. In word spotting, the goal is to find all instances of a query word in a dataset of images. In recognition, the goal is to recognize the content of the word image, usually aided by a dictionary or lexicon. We describe an approach in which both word images and text strings are embedded in a common vectorial subspace. This is achieved by a combination of label embedding and attributes learning, and a common subspace regression. In this subspace, images and strings that represent the same word are close together, allowing one to cast recognition and retrieval tasks as a nearest neighbor problem. Contrary to most other existing methods, our representation has a fixed length, is low dimensional, and is very fast to compute and, especially, to compare. We test our approach on four public datasets of both handwritten documents and natural images showing results comparable or better than the state-of-the-art on spotting and recognition tasks.

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.

In this paper we deal with the problem of recognizing handwritten shapes. We present a new deformable feature extraction method that adapts to the shape to be described, dealing in this way with the variability introduced in the handwriting domain. It consists in a selection of the regions that best define the shape to be described, followed by the computation of histograms of oriented gradients-based features over these points. Our results significantly outperform other descriptors in the literature for the task of hand-drawn shape recognition and handwritten word retrieval.

This report presents the final results of the ICDAR 2013 Robust Reading Competition. The competition is structured in three Challenges addressing text extraction in different application domains, namely born-digital images, real scene images and real-scene videos. The Challenges are organised around specific tasks covering text localisation, text segmentation and word recognition. The competition took place in the first quarter of 2013, and received a total of 42 submissions over the different tasks offered. This report describes the datasets and ground truth specification, details the performance evaluation protocols used and presents the final results along with a brief summary of the participating methods.

In this paper we propose an unsupervised segmentation-free method for word spotting in document images. Documents are represented with a grid of HOG descriptors, and a sliding window approach is used to locate the document regions that are most similar to the query. We use the Exemplar SVM framework to produce a better representation of the query in an unsupervised way. Finally, the document descriptors are precomputed and compressed with Product Quantization. This offers two advantages: first, a large number of documents can be kept in RAM memory at the same time. Second, the sliding window becomes significantly faster since distances between quantized HOG descriptors can be precomputed. Our results significantly outperform other segmentation-free methods in the literature, both in accuracy and in speed and memory usage.

In this paper we propose an approach for word spotting in handwritten document images. We state the problem from a focused retrieval perspective, i.e. locating instances of a query word in a large scale dataset of digitized manuscripts. We combine two approaches, namely one based on word segmentation and another one segmentation-free. The first approach uses a hashing strategy to coarsely prune word images that are unlikely to be instances of the query word. This process is fast but has a low precision due to the errors introduced in the segmentation step. The regions containing candidate words are sent to the second process based on a state of the art technique from the visual object detection field. This discriminative model represents the appearance of the query word and computes a similarity score. In this way we propose a coarse-to-fine approach achieving a compromise between efficiency and accuracy. The validation of the model is shown using a collection of old handwritten manuscripts. We appreciate a substantial improvement in terms of precision regarding the previous proposed method with a low computational cost increase.

This paper presents a methodology for shape recognition that focuses on dealing with the difficult problem of large deformations. The proposed methodology consists in a novel feature extraction technique, which uses a non-rigid representation adaptable to the shape. This technique employs a deformable grid based on the computation of geometrical centroids that follows a region partitioning algorithm. Then, a feature vector is extracted by computing pixel density measures around these geometrical centroids. The result is a shape descriptor that adapts its representation to the given shape and encodes the pixel density distribution. The validity of the method when dealing with large deformations has been experimentally shown over datasets composed of handwritten shapes. It has been applied to signature verification and shape recognition tasks demonstrating high accuracy and low computational cost.