Publications by Costantino Grana

The diffusion of powerful mobile devices has posed the basis for new applications implementing on the devices (which are embedded devices) sophisticated computer vision and pattern recognition algorithms. This paper describes the implementation of a complete system for automatic recognition of places localized on a map through the recognition of significant signs by means of the camera of a mobile device (smartphone, tablet, etc.). The paper proposes a novel classification algorithm based on the innovative use of bag-of-words on ORB features. The recognition is achieved using a simple yet effective search scheme which exploits GPS localization to limit the possible matches. This simple solution brings several advantages, such as the speed also on limited-resource devices, the usability also with limited training samples and the easiness of adapting to new training samples and classes. The overall architecture of the system is based on a REST-JSON client-server architecture. The experimental results have been conducted in a real scenario and evaluating the different parameters which influence the performance.

Common techniques represent images by quantizing local descriptors and summarizing their distribution in a histogram. In this paper we propose to employ a parametric description and compare its capabilities to histogram based approaches. We use the multivariate Gaussian distribution, applied over the SIFT descriptors, extracted with dense sampling on a spatial pyramid. Every distribution is converted to a high-dimensional descriptor, by concatenating the mean vector and the projection of the covariance matrix on the Euclidean space tangent to the Riemannian manifold. Experiments on Caltech-101 and ImageCLEF2011 are performed using the Stochastic Gradient Descent solver, which allows to deal with large scale datasets and high dimensional feature spaces.

In this paper we propose a large-scale Image annotation system for the Scalable Concept Image Annotation task. For each concept to be detected a separated classifier is built using the provided textual annotation. Images are represented as a Multivariate Gaussian distribution of a set of local features extracted over a dense regular grid. Textual analysis, on the web pages containing training images, is performed to retrieve a relevant set of samples for learning each concept classifier. An online SVMs solver based on Stochastic Gradient Descent is used to manage the large amount of training data. Experimental results show that the combination of different kind of local features encoded with our strategy achieves very competitive performance both in terms of mAP and mean F-measure.

In this paper, we suggest an interaction model designed to fit users' expectations in front of an image retrieval system. A lightweight relevance feedback strategy, working directly on the 2D projection of image features, allows the user to spatially navigate the media collection maintaining the real-time constraint. A preliminary evaluation of this relevance feedback strategy shows good performance compared with other known approaches.

For this demo, we present a set of stark applications designed to evaluate the performance of a color similarity retrieval system against human operators performance in the same tasks. The proposed series of tests give some interesting insights about the perception of color classes and the reliability of manual annotation in the fashion context. © 2012 Springer-Verlag.

Color signatures, histograms and bag of colors are basic and effective strategies for describing the color content of images, for retrieving images by their color appearance or providing color annotation. In some domains, colors assume a specific meaning for users and the color-based classification and retrieval should mirror the initial suggestions given by users in the training set. For instance in fashion world, the names given to the dominant color of a garment or a dress reflect the fashion dictact and not an uniform division of the color space.In this paper we propose a general approach to implement color signature as a trained bag of words, defined on the basis of user defined color classes. The novel Class-based Color Bag of Words is a easy computable bag of words of color, constructed following an approach similar to the Median Cut algorithm, but biased by color distribution in the trained classes. Moreover, to dramatically reduce the computational effort we propose 3D integral histograms, a 3D extension of integral images, easily extensible for many histogram-based signature in 3D color space. Several comparisons in large fashion datasets confirm the discriminant power of this signature.

In this paper we propose a color-based approach for skin detection and interest garment selection aimed at an automatic segmentation of pieces of clothing. For both purposes, the color description is extracted by an iterative energy minimization approach and an automatic initialization strategy is proposed by learning geometric constraints and shape cues. Experiments confirms the good performance of this technique both in the context of skin removal and in the context of classification of garments.

Multimedia technologies have recently created the conditions for a true revolution in the Cultural Heritage domain, particularly in reference to the study, exploitation, and fruition of artistic works. New opportunities are arising for researchers in the field of multimedia to share their research results with people coming from the field of art and culture, and viceversa. This paper gathers together opinions and ideas shared during the final discussion session at the 1st International Workshop on Multimedia for Cultural Heritage, as a summary of the problems and possible directions to solve to them.

In this paper we present a novel algorithm to synthesize an optimal decision tree from OR-decision tables, an extension of standard decision tables, complete with the formal proof of optimality and computational cost analysis. As many problems which require to recognize particular patterns can be modeled with this formalism, we select two common binary image processing algorithms, namely connected components labeling and thinning, to show how these can be represented with decision tables, and the benets of their implementation as optimal decision trees in terms of reduced memory accesses. Experiments are reported, to show the computational time improvements over state of the art implementations.