You are here

Stefano Messelodi

Head of Unit
  • Phone: 0461314509
  • FBK Povo
Short bio

Stefano Messelodi was born in Arco (Italy) in 1961. He graduated in computer science from the University of Milan (Italy).  Since 1986 he is working in FBK (ITC-irst), Trento, Italy, where he coordinated the Technologies of Vision research unit till 2012. His research interests include text localization in scene, semantic image labelling and dynamic scene understanding. He served as a reviewer for several journals and conferences. He is a member of IEEE Society and International Association for Pattern Recognition.
He currently coordinates the research unit TeV.

Research interests
people re-identification text localization in scene document image analysis traffic analysis
Publications
  1. S. Messelodi; C.M. Modena,
    A Computer Vision System for Traffic Accident Risk Measurement: A Case Study,
    Abstract - A reliable estimation of the safety level of the roads is a valuable tool for detecting critical points in the road infrastructure, planning and implement countermeasures, and evaluating their impact on the traffic. A method for the computation of the accident risk is proposed, which is based on microscopic traffic data collected automatically by a video-based monitoring system, i.e. class, speed, and tra jectory of each single road-user. The benefit of the proposed method is twofold: the risk level is computed without statistics on past accidents, and its computation is fully automated, i.e. it does not require a manual collection of traffic data. The paper presents the definition of the proposed risk index and
    describes its application to a real case: the evaluation of the accident risk at an urban intersection, before and after the reorganization of its geometry. The proposed risk index, although based only on those parameters that are automatically measurable, seems to reflect the expectation of traffic experts in evaluating the impact of intervention to improve the safety level of the intersection.
    ,
    2005
  2. C. Corridori; D. Giordani; P. Lombardi; S. Messelodi; C. M. Modena; M. Zanin,
    An in-vehicle vision system for dangerous situation detection,
    Proceedings,
    Perugia,
    Morlacchi Editore,
    2004
    , (Conferenza Italiana sui Sistemi Intelligenti 2004 – 2° Convegno del Gruppo Italiano Ricercatori in Pattern Recognition GIRPR-04,
    Perugia, Italy,
    from 09/15/2004 to 09/15/2004)
  3. C. Andreatta; M. Lecca; S. Messelodi,
    MEMORI,
    RIAO 2004. Coupling approaches, coupling media and coupling languages for information retrieval,
    2004
    , (RIAO 2004. Coupling approaches, coupling media and coupling languages for information retrieval,
    26/04/2004)
  4. Qi. Xu; R. Brunelli; S. Messelodi; J. Zhang; M. Li,
    Image Coherence Based Adaptive Sampling for Image Synthesis,
    Computational Science and its Applications - ICCSA 2004,
    Springer,
    vol.3044/2004,
    2004
    , pp. 693-
    702
    , (International Conference on Computational Science and its Applications - ICCSA 2004,
    Assisi, Italy,
    05/2004)
  5. Q. Xu; L. Ma; M. Li; W. Wang; J. Cai; R. Brunelli; S. Messelodi,
    Fuzzy weighted average filtering for mixture noises,
    2004
    , pp. 18-
    21
    , (Third International Conference on Image and Graphics (ICIG'04),
    Hong Kong, China,
    12/18/2004 - 12/20/2004)
  6. S. Messelodi; C. M. Modena; M. Zanin,
    A computer vision system for the detection and classification of vehicles at urban road intersections,
    This paper presents a real-time vision system to compute traffic parameters by analyzing monocular image sequences from pole-mounted video cameras at urban crossroads.
    The system is flexible with respect to road geometry and camera position, permitting data collection from several monitored crosses. It uses a combination of segmentation and motion information to localize and track moving objects on the road plane, utilizing background subtraction and a feature-based tracking methodology. For each detected vehicle, the system is able to describe its path, to estimate its speed and to classify it into seven categories. The classification task relies on a model-based matching technique refined by a feature-based one.
    Experimental results demonstrate robust, real-time vehicle detection, tracking and classification over several hours of videos taken under different illumination conditions. The system is presently under trial in Trento, a one hundred thousand people town in northern Italy
    ,
    2004
  7. G. Cattoni; S. Messelodi; C. M. Modena,
    Vision-based bicycle/motorcycle classification with Support Vector Machines,
    Classification of vehicles plays an important role in a traffic monitoring system. In this paper we present a feature-based classifier, which can distinguish bicycles from motorcycles in real world traffic scenes.
    Basically, the algorithm extracts some visual features focusing on the region corresponding to the wheels of vehicle. It splits the problem into two sub-cases depending on the computed motion direction. The classification is performed by means of a non-linear Support Vector Machine. Tests lead to a successful classification rate of 93% on video sequences taken from different road junctions in an urban environment
    ,
    2004
  8. C. Andreatta; M. Lecca; S. Messelodi,
    Memory based object recognition in images,
    MEMORI is a system for the detection and recognition of objects, stored in a database, within digital images taken as an input. It aims to provide a content-based indexing scheme which exploits the semantic information contained in the images. Object detection is achieved by segmenting the input image using color and textural information, and grouping the obtained regions by analyzing their adjacency relationships and their visual similarity to the objects in the database. The database is the memory of the system and consists of objects depicted by a set of different views, each of them described by a feature vector encoding color, texture and shape information. The same description is computed for the region groups candidated to represent an object and is compared, by an object classifier, with the database objects. The final result is a number of image regions each associated to one or more object classes with a confidence score,
    2004
  9. Michele Zanin; Stefano Messelodi; Carla Maria Modena,
    Mantova, Italy,
    17/09/2003 - 19/09/2003,
  10. Stefano Messelodi; Carla Maria Modena,
    Extraction of Polygonal Frames from Color Documents for Page Decomposition,
    Graphic accents are often used in the design of complex documents in order to emphasize particular information. Words or illustrations are surrounded by a border line or highlighted by means of a colored background. This paper presents a method for the automatic extraction of document layout items, called {\em frames}, having polygonal shape and/or a uniformly colored background. As frames break the normal text flow, frame detection is a fundamental step of the document layout analysis in a document understanding system. The presented method relies on a color region growing algorithm and on straight edges extractor. The shape analysis of the obtained regions permits to localize the frames with their attributes. In order to reduce computation time and to return only specific patterns, the method exploits information about a model of the frames to be detected such as shape, skew or size, possibly supplied by the user or depending on the specific document class. The presented algorithm is assessed on a page databases containing more than 675 framed items. The evaluation is based on a novel tree matching method that takes into account the frame hierarchy and their shape,
    2003

Pages