Lorenzo Sarti

2005 Special Issue: Recursive neural networks for processing graphs with labelled edges: theory and applications

In this paper, we introduce a new recursive neural network model able to process directed acyclic graphs with labelled edges. The model uses a state transition function which considers the edge labels and is independent both from the number and the order of the children of each node. The computational capabilities of the new recursive architecture are assessed. Moreover, in order to test the proposed architecture on a practical challenging application, the problem of object detection in images is also addressed. In fact, the localization of target objects is a preliminary step in any recognition system. The proposed technique is general and can be applied in different detection systems, since it does not exploit any a priori knowledge on the particular problem. Some experiments on face detection, carried out on scenes acquired by an indoor camera, are reported, showing very promising results.

Recursive neural networks learn to localize faces

Localizing faces in images is a difficult task, and represents the first step towards the solution of the face recognition problem. Moreover, devising an effective face detection method can provide some suggestions to solve similar object and pattern detection problems. This paper presents a novel approach to the solution of the face localization problem using Recursive neural networks (RNNs). The proposed method assumes a graph-based representation of images that combines structural and symbolic visual features. Such graphs are then processed by RNNs, in order to establish the possible presence and the position of faces inside the image. A novel RNN model that can deal with graphs with labeled edges has been also exploited. Some experiments on snapshots from video sequences are reported, showing very promising results.

Recursive processing of cyclic graphs

Recursive neural networks are a powerful tool for processing structured data. According to the recursive learning paradigm, the input information consists of directed positional acyclic graphs (DPAGs). In fact, recursive networks are fed following the partial order defined by the links of the graph. Unfortunately, the hypothesis of processing DPAGs is sometimes too restrictive, being the nature of some real-world problems intrinsically cyclic. In this paper, a methodology is proposed, which allows us to process any cyclic directed graph. Therefore, the computational power of recursive networks is definitely established, also clarifying the underlying limitations of the model.

Graph matching using random walks

In this paper, we propose a graph matching algorithm which uses random walks to compute topological features for each node, in order to identify candidate pairs of corresponding nodes in the two graphs. The algorithm automatically adapts the number of topological features required to determine the exact match among the nodes. Even if the proposed technique is not guaranteed to provide an exact solution for all graphs, the experiments on a benchmark dataset show that it can outperform other state of the art algorithms with respect to the computational requirements. In fact, the proposed algorithm is polynomial in the number of graph nodes.

A recursive neural network model for processing directed acyclic graphs with labeled edges

The recursive paradigm extends the neural network processing and learning algorithms to deal with structured inputs. In particular, recursive neural network (RNN) models have been proposed to process information coded as directed positional acyclic graphs (DPAGs) whose maximum node outdegree is known a priori. Unfortunately, the hypothesis of processing DPAGs having a given maximum node outdegree is sometimes too restrictive, being the nature of some real-world problems intrinsically disordered. In many applications the node outdegrees can vary considerably among the nodes in the graph, it may be unnatural to define a position for each child of a given node, and it may be necessary to prune some edges to reduce the number of the network parameters, which is proportional to the maximum node outdegree. In this paper, we proposed a new recursive neural network model which allows us to process directed acyclic graphs (DAGs) with labeled edges, relaxing the positional constraint and the correlated maximum outdegree limit. The effectiveness of the new scheme is experimentally tested on an image classification task. The results show that the new RNN model outperforms the standard RNN architecture, also allowing us to use a smaller number of free parameters.

A template-based approach to automatic face enhancement

This paper presents Visual ENhancement of USers (VENUS), a system able to automatically enhance male and female frontal facial images exploiting a database of celebrities as reference patterns for attractiveness. Each face is represented by a set of landmark points that can be manually selected or automatically localized using active shape models. The faces can be compared remapping the landmarks by means of Catmull–Rom splines, a class of interpolating splines particularly useful to extract shape-based representations. Given the input image, its landmarks are compared against the known beauty templates and moved towards the K-nearest ones by 2D image warping. The VENUS performances have been evaluated by 20 volunteers on a set of images collected during the Festival of Creativity, held in Florence, Italy, on October 2007. The experiments show that the 73.9% of the beautified faces are more attractive than the original pictures.

An eye detection system based on neural autoassociators

Automatic eye tracking is a challenging task, with numerous applications in biometrics, security, intelligent human–computer interfaces, and drivers sleepiness detection systems. Eye localization and extraction is, therefore, the first step to the solution of such problems. In this paper, we present a new method, based on neural autoassociators, to solve the problem of detecting eyes from a facial image. A subset of the AR Database, collecting individuals both with or without glasses and with open or closed eyes, has been used for experiments and benchmarking. Preliminary experimental results are very promising and demonstrate the efficiency of the proposed eye localization system.

Recursive neural networks for object detection

In this paper, a new recursive neural network model, able to process directed acyclic graphs with labeled edges, is introduced, in order to address the problem of object detection in images. In fact, the detection is a preliminary step in any object recognition system. The proposed method assumes a graph-based representation of images, that combines both spatial and visual features. In particular, after segmentation, an edge between two nodes stands for the adjacency relationship of two homogeneous regions, the edge label collects information on their relative positions, whereas node labels contain visual and geometric information on each region (area, color, texture, etc.). Such graphs are then processed by the recursive model in order to determine the eventual presence and the position of objects inside the image. Some experiments on face detection, carried out on scenes acquired by an indoor camera, are reported, showing very promising results. The proposed technique is general and can be applied in different object detection systems, since it does not include any a priori knowledge on the particular problem.

A Neural Network Approach to Similarity Learning

This paper presents a novel neural network model, called similarity neural network (SNN), designed to learn similarity measures for pairs of patterns. The model guarantees to compute a non negative and symmetric measure, and shows good generalization capabilities even if a very small set of supervised examples is used for training. Preliminary experiments, carried out on some UCI datasets, are presented, showing promising results.

Face localization with recursive neural networks

Recognizing a particular face in a complex image or in a video sequence, which the humans can simply accomplish using contextual information, is a difficult task for an automatic recognizer. Moreover, the face recognition problem is usually solved having assumed that the face was previously localized, often via heuristics based on prototypes of the whole face or significant details. In this paper, we propose a novel approach to the solution of the face localization problem using recursive neural networks. In particular, the proposed approach assumes a graph–based representation of images that combines structural and sub–symbolic visual features. Such graphs are then processed by recursive neural networks, in order to establish the eventual presence and the position of the faces inside the image. Some preliminary experiments on snapshots from video sequences are reported, showing very promising results.