MMS-ER3D 2015 Abstracts


Full Papers
Paper Nr: 1
Title:

Accurate Detection and Visualization of 3D Shape Deformation by using Multiple Projectors

Authors:

Masayasu Yoshigi, Fumihiko Sakaue and Jun Sato

Abstract: In this paper, we propose a method for detecting the deformation of object shape by using multiple projectors. In this method, a set of specially coded patterns are projected onto a target object from multiple projectors. Then, if the target object is not deformed, the object is illuminated by plain white color, and if the object is deformed, it is illuminated by radical colors. Thus, we can visualize and detect the deformation of object just by projecting lights from multiple projectors. The proposed method uses the disparities of multiple projectors, and thus, we do not any complicated method for detecting object shape deformation. In addition, we utilize image super-resolution technique for object deformation visualization, so that we can visualize extremely small deformation easily.

Paper Nr: 2
Title:

Maritime Targets Detection from Ground Cameras Exploiting Semi-supervised Machine Learning

Authors:

Eftychios Protopapadakis, Konstantinos Makantasis and Nikolaos Doulamis

Abstract: This paper presents a vision-based system for maritime surveillance, using moving PTZ cameras. The proposed methodology fuses a visual attention method that exploits low-level image features appropriately selected for maritime environment, with appropriate tracker. Such features require no assumptions about environmental nor visual conditions. The offline initialization is based on large graph semi-supervised technique in order to minimize user’s effort. System’s performance was evaluated with videos from cameras placed at Limassol port and Venetian port of Chania. Results suggest high detection ability, despite dynamically changing visual conditions and different kinds of vessels, all in real time.

Paper Nr: 3
Title:

A 3D Feature for Building Segmentation based on Shape-from-Shading

Authors:

Dimitrios Konstantinidis, Vasileios Argyriou, Tania Stathaki and Nikos Grammalidis

Abstract: An important cue that can assist towards an accurate building detection and segmentation is 3D information. Because of their height, buildings can easily be distinguished from the ground and small objects, allowing for their successful segmentation. Unfortunately, 3D knowledge is not always available, but there are ways to infer 3D information from 2D images. Shape-from-shading techniques extract height and surface normal information from a single 2D image by taking into consideration knowledge about illumination, reflectance and shape. In this paper, a novel feature is proposed that can describe the 3D information of reconstructed images based on a shape-from-shading technique in order to successfully acquire building boundaries. The results are promising and show that such a 3D feature can significantly assist in a correct building boundary detection and segmentation.

Paper Nr: 4
Title:

Prior Knowledge About Camera Motion for Outlier Removal in Feature Matching

Authors:

Elisavet K. Stathopoulou, Ronny Hänsch and Olaf Hellwich

Abstract: The search of corresponding points in between images of the same scene is a well known problem in many computer vision applications. In particular most structure from motion techniques depend heavily on the correct estimation of corresponding image points. Most commonly used approaches make neither assumptions about the 3D scene nor about the relative positions of the cameras and model both as completely unknown. This general model results in a brute force comparison of all keypoints in one image to all points in all other images. In reality this model is often far too general because coarse prior knowledge about the cameras is often available. For example, several imaging systems are equipped with positioning devices which deliver pose information of the camera. Such information can be used to constrain the subsequent point matching not only to reduce the computational load, but also to increase the accuracy of path estimation and 3D reconstruction. This study presents Guided Matching as a new matching algorithm towards this direction. The proposed algorithm outperforms brute force matching in speed as well as number and accuracy of correspondences, given well estimated priors.

Paper Nr: 5
Title:

Event-complementing Online Human Life Summarization based on Social Latent Semantic Analysis

Authors:

Klimis S. Ntalianis and Anastasios D. Doulamis

Abstract: In this paper, online human life summarization is performed, based on multimedia content, published on social media. The life summaries are also automatically annotated with events, persons, places etc. Towards this direction, initially a content preparation module is activated that includes an intelligent wrapper. The content preparation module scans social networks, extracts their pages and segments them into tokens, in an unsupervised way. Next multimedia content is kept and it is associated to its respective metadata. In the following step, a novel ranking mechanism puts multimedia content in order of importance based on usercontent interactions. Finally the event-complementing summarization module produces a meaningful annotated video clip, based on a spectral visual clustering technique and the innovative Social Latent Semantic Analysis algorithm. Experimental results illustrate the promising performance of the proposed architecture and set some foundations for future research.

Paper Nr: 6
Title:

HDR Imaging for Enchancing People Detection and Tracking in Indoor Environments

Authors:

Panagiotis Agrafiotis, Elisavet K. Stathopoulou, Andreas Georgopoulos and Anastasios Doulamis

Abstract: Videos and image sequences of indoor environments with challenging illumination conditions often capture either brightly lit or dark scenes where every single exposure may contain overexposed and/or underexposed regions. High Dynamic Range (HDR) images contain information that standard dynamic range ones, often mentioned also as low dynamic range images (SDR/LDR) cannot capture. This paper investigates the contribution of HDR imaging in people detection and tracking systems. In order to evaluate this contribution of the HDR imaging in the accuracy and robustness of pedestrian detection and tracking in challenging indoor visual conditions, two state of the art trackers of different complexity were implemented. To this direction data were collected taking into account the requirements and real-life indoor scenarios and HDR frames were produced. The algorithms were applied to the SDR data and their corresponding HDR data and were compared and evaluated for their robustness and accuracy in terms of precision and recall. Results show that that the use of HDR images enhances the performance of the detection and tracking scheme, making it robust and more reliable.

Paper Nr: 7
Title:

A 4D Virtual/Augmented Reality Viewer Exploiting Unstructured Web-based Image Data

Authors:

Anastasios Doulamis, Nikolaos Doulamis, Konstantinos Makantasis and Michael Klein

Abstract: Outdoor large-scale cultural sites are mostly sensitive to environmental, natural and human made factors, implying an imminent need for a spatio-temporal assessment to identify regions of potential cultural interest (material degradation, structuring, conservation). Thus, 4D modelling (3D plus the time) is ideally required for preservation and assessment of outdoor large scale cultural sites, which is currently implemented as a simple aggregation of 3D digital models at different time. However, it is difficult to implement temporal 3D modelling for many time instances using conventional capturing tools since we need high financial effort and computational complexity in acquiring a set of the most suitable image data. One way to address this, is to exploit the huge amount of images distributing over visual hosting repositories, such as flickr and picasa. These visual data, nevertheless, are loosely structured and thus no appropriate for 3D modelling. For this reason, a new content-based filtering mechanism should be implemented so as to rank (filter) images according to their contribution to the 3D reconstruction process and discards image outliers that can either confuse or delay the 3D reconstruction process. Then, we proceed to the implementation of a virtual/augmented reality which allows the cultural heritage actors to temporally assess cultural objects of interest and assists conservators to check how restoration methods affect an object or how materials decay through time. The proposed system has been developed and evaluated using real-life data and outdoor sites.

Paper Nr: 8
Title:

A Novel Human Interaction Game-like Application to Learn, Perform and Evaluate Modern Contemporary Singing - "Human Beat Box"

Authors:

S. K. Al Kork, D. Uğurca, C. Şahin, P. Chawah, L. Buchman, M. Adda-Decker, K. Xu, B. Denby, P. Roussel, B. Picart, S. Dupont, F. Tsalakanidou, A. Kitsikidis, F. M. Dagnino, M. Ott, F. Pozzi, M. Stone and E. Yilmaz

Abstract: The paper presents an interactive game-like application to learn, perform and evaluate modern contemporary singing. The Human Beat Box (HBB) is being used as a case study. The game consists of two main modules. A sensor module that consists of a portable helmet based system containing an ultrasonic (US) transducer to capture tongue movements, a video camera for the lips, Kinect camera for face gestures, and a microphone for sound. A 3D environment game module is used to visualize a 3D recording studio as game world with all of its unique elements like guitars, mixer, amplifier, speakers and a microphone in front of the 3D avatar to simulate the recording ambience. The game also features a 2D virtual tutor to help the learner by giving oral and written feedback during the game. He also gives feedbacks during the practice session to improve the student’s performance. The game is still at its early stages of development and it is been tested using simple HBB plosive sounds for percussion such as “PTK”.

Paper Nr: 9
Title:

Novel 3D Game-like Applications Driven by Body Interactions for Learning Specific Forms of Intangible Cultural Heritage

Authors:

E. Yilmaz, D. Uğurca, C. Şahin, F. M. Dagnino, M. Ott, F. Pozzi, K. Dimitropoulos, F. Tsalakanidou, A. Kitsikidis, S. K. Al Kork, K. Xu, B. Denby, P. Roussel, P. Chawah, L. Buchman, M. Adda-Decker, S. Dupont, B. Picart, J. Tilmanne, M. Alivizatou, L. Hadjileontiadis, V. Charisis, A. Glushkova, C. Volioti, A. Manitsaris, E. Hemery, F. Moutarde and N. Grammalidis

Abstract: The main objective of the EU FP7 ICT i-Treasures project is to build a public and expandable platform to enable learning and transmission of rare know-how of intangible cultural heritage. A core part of this platform consists of game-like applications able to support teaching and learning processes in the ICH field. We have designed and developed four game-like applications (for Human Beat Box singing, Tsamiko dancing, pottery making and contemporary music composition), each corresponding to one of the ICH use cases of i-Treasures project. A first preliminary version of these applications is currently available for further validation, evaluation and demonstration within the project. We have encountered a number of issues, most of which derive from the peculiarities of the ICH domains addressed by the project, and many have already been resolved/ The evaluation results are expected to lead to further optimization of these games.

Short Papers
Paper Nr: 10
Title:

Lightweight Computer Vision Methods for Traffic Flow Monitoring on Low Power Embedded Sensors

Authors:

Massimo Magrini, Davide Moroni, Gabriele Pieri and Ovidio Salvetti

Abstract: Nowadays pervasive monitoring of traffic flows in urban environment is a topic of great relevance, since the information it is possible to gather may be exploited for a more efficient and sustainable mobility. In this paper, we address the use of smart cameras for assessing the level of service of roads and early detect possible congestion. In particular, we devise a lightweight method that is suitable for use on low power and low cost sensors, resulting in a scalable and sustainable approach to flow monitoring over large areas. We also present the current prototype of an ad hoc device we designed and report experimental results obtained during a field test.