Abstract: In this paper we present a multipage administrative document image retrieval system based on textual and visual representations of document pages. Individual pages are represented by textual or visual information using a bag-of-words framework. Different fusion strategies are evaluated which allow the system to perform multipage document retrieval on the basis of a single page retrieval system. Results are reported on a large dataset of document images sampled from a banking workflow.

Abstract: Today, robust learners trained in a real supervised machine learning application should count with a rich collection of positive and negative examples. Although in many applications, it is not difficult to obtain huge amount of data, labeling those data can be a very expensive process, especially when dealing with data of high variability and complexity. A good example of such cases are data from medical imaging applications where annotating anomalies like tumors, polyps, atherosclerotic plaque or informative frames in wireless endoscopy need highly trained experts. Building a representative set of training data from medical videos (e.g. Wireless Capsule Endoscopy) means that thousands of frames to be labeled by an expert. It is quite normal that data in new videos come different and thus are not represented by the training set. In this paper, we review the main approaches on active learning and illustrate how active learning can help to reduce expert effort in constructing the training sets. We show that applying active learning criteria, the number of human interventions can be significantly reduced. The proposed system allows the annotation of informative/non-informative frames of Wireless Capsule Endoscopy video containing more than 30000 frames each one with less than 100 expert ”clicks”.

Abstract: This paper addresses visual object perception applied to mobile robotics. Being able to perceive household objects in unstructured environments is a key capability in order to make robots suitable to perform complex tasks in home environments. However, finding a solution for this task is daunting: it requires the ability to handle the variability in image formation in a moving camera with tight time constraints. The paper brings to attention some of the issues with applying three state of the art object recognition and detection methods in a mobile robotics scenario, and proposes methods to deal with windowing/segmentation. Thus, this work aims at evaluating the state-of-the-art in object perception in an attempt to develop a lightweight solution for mobile robotics use/research in typical indoor settings.

Abstract: Many natural image sets, depicting objects whose appearance is changing due to motion, pose or light variations, can be considered samples of a low-dimension nonlinear manifold embedded in the high-dimensional observation space (the space of all possible images). The main contribution of our work is represented by a Supervised Laplacian Eigemaps (S-LE) algorithm, which exploits the class label information for mapping the original data in the embedded space. Our proposed approach benefits from two important properties: i) it is discriminative, and ii) it adaptively selects the neighbors of a sample without using any predefined neighborhood size. Experiments were conducted on four face databases and the results demonstrate that the proposed algorithm significantly outperforms many linear and non-linear embedding techniques. Although we've focused on the face recognition problem, the proposed approach could also be extended to other category of objects characterized by large variance in their appearance.

Abstract: IF= 2.61
IF=2.61 (2010)
In this paper we introduce a novel supervised manifold learning technique called Supervised Laplacian Eigenmaps (S-LE), which makes use of class label information to guide the procedure of non-linear dimensionality reduction by adopting the large margin concept. The graph Laplacian is split into two components: within-class graph and between-class graph to better characterize the discriminant property of the data. Our approach has two important characteristics: (i) it adaptively estimates the local neighborhood surrounding each sample based on data density and similarity and (ii) the objective function simultaneously maximizes the local margin between heterogeneous samples and pushes the homogeneous samples closer to each other.
Our approach has been tested on several challenging face databases and it has been conveniently compared with other linear and non-linear techniques, demonstrating its superiority. Although we have concentrated in this paper on the face recognition problem, the proposed approach could also be applied to other category of objects characterized by large variations in their appearance (such as hand or body pose, for instance.

Abstract: Different methodologies of uniform sampling over the rotation group, SO(3), for building unbiased 2D shape models from 3D objects are introduced and reviewed in this chapter. State-of-the-art non uniform sampling approaches are discussed, and uniform sampling methods using Euler angles and quaternions are introduced. Moreover, since presented work is oriented to model building applications, it is not limited to general discrete methods to obtain uniform 3D rotations, but also from a continuous point of view in the case of Procrustes Analysis.

Abstract: The ability to automatically detect and recognize objects in unconstrained images is becoming more and more critical: from security systems and autonomous robots, to smart phones and augmented reality, intelligent devices need to understand the meaning of images as a composition of semantic objects. This Thesis tackles the problem of fast object detection based on template models. Detection consists of searching for an object in an image by evaluating the similarity between a template model and an image region at each possible location and scale. In this work, we show that using a template model representation based on a multiple resolution hierarchy is an optimal choice that can lead to excellent detection accuracy and fast computation. We implement two different approaches that make use of a hierarchy of multiresolution models: a multiresolution cascade and a coarse-to-fine search. Also, we extend the coarse-to-fine search by introducing a deformable part-based model that achieves state-of-the-art results together with a very reduced computational cost. Finally, we specialize our approach to the challenging task of pedestrian detection from moving vehicles and show that the overall quality of the system outperforms previous works in terms of speed and accuracy.