toggle visibility Search & Display Options

Select All    Deselect All
 |   | 
Details
  Records Links
Author Victor Ponce; Sergio Escalera; Marc Perez; Oriol Janes; Xavier Baro edit  doi
openurl 
  Title Non-Verbal Communication Analysis in Victim-Offender Mediations Type Journal Article
  Year 2015 Publication Pattern Recognition Letters Abbreviated Journal PRL  
  Volume 67 Issue 1 Pages 19-27  
  Keywords Victim–Offender Mediation; Multi-modal human behavior analysis; Face and gesture recognition; Social signal processing; Computer vision; Machine learning  
  Abstract We present a non-invasive ambient intelligence framework for the semi-automatic analysis of non-verbal communication applied to the restorative justice field. We propose the use of computer vision and social signal processing technologies in real scenarios of Victim–Offender Mediations, applying feature extraction techniques to multi-modal audio-RGB-depth data. We compute a set of behavioral indicators that define communicative cues from the fields of psychology and observational methodology. We test our methodology on data captured in real Victim–Offender Mediation sessions in Catalonia. We define the ground truth based on expert opinions when annotating the observed social responses. Using different state of the art binary classification approaches, our system achieves recognition accuracies of 86% when predicting satisfaction, and 79% when predicting both agreement and receptivity. Applying a regression strategy, we obtain a mean deviation for the predictions between 0.5 and 0.7 in the range [1–5] for the computed social signals.  
  Address  
  Corporate Author Thesis  
  Publisher Place of Publication Editor  
  Language Summary Language Original Title  
  Series Editor Series Title Abbreviated Series Title  
  Series Volume Series Issue Edition  
  ISSN (up) ISBN Medium  
  Area Expedition Conference  
  Notes HuPBA;MV Approved no  
  Call Number Admin @ si @ PEP2015 Serial 2583  
Permanent link to this record
 

 
Author Frederic Sampedro; Sergio Escalera; Anna Domenech; Ignasi Carrio edit  doi
openurl 
  Title Automatic Tumor Volume Segmentation in Whole-Body PET/CT Scans: A Supervised Learning Approach Source Type Journal Article
  Year 2015 Publication Journal of Medical Imaging and Health Informatics Abbreviated Journal JMIHI  
  Volume 5 Issue 2 Pages 192-201  
  Keywords CONTEXTUAL CLASSIFICATION; PET/CT; SUPERVISED LEARNING; TUMOR SEGMENTATION; WHOLE BODY  
  Abstract Whole-body 3D PET/CT tumoral volume segmentation provides relevant diagnostic and prognostic information in clinical oncology and nuclear medicine. Carrying out this procedure manually by a medical expert is time consuming and suffers from inter- and intra-observer variabilities. In this paper, a completely automatic approach to this task is presented. First, the problem is stated and described both in clinical and technological terms. Then, a novel supervised learning segmentation framework is introduced. The segmentation by learning approach is defined within a Cascade of Adaboost classifiers and a 3D contextual proposal of Multiscale Stacked Sequential Learning. Segmentation accuracy results on 200 Breast Cancer whole body PET/CT volumes show mean 49% sensitivity, 99.993% specificity and 39% Jaccard overlap Index, which represent good performance results both at the clinical and technological level.  
  Address  
  Corporate Author Thesis  
  Publisher Place of Publication Editor  
  Language Summary Language Original Title  
  Series Editor Series Title Abbreviated Series Title  
  Series Volume Series Issue Edition  
  ISSN (up) ISBN Medium  
  Area Expedition Conference  
  Notes HuPBA;MILAB Approved no  
  Call Number Admin @ si @ SED2015 Serial 2584  
Permanent link to this record
 

 
Author Frederic Sampedro; Anna Domenech; Sergio Escalera edit  url
doi  openurl
  Title Static and dynamic computational cancer spread quantification in whole body FDG-PET/CT scans Type Journal Article
  Year 2014 Publication Journal of Medical Imaging and Health Informatics Abbreviated Journal JMIHI  
  Volume 4 Issue 6 Pages 825-831  
  Keywords CANCER SPREAD; COMPUTER AIDED DIAGNOSIS; MEDICAL IMAGING; TUMOR QUANTIFICATION  
  Abstract In this work we address the computational cancer spread quantification scenario in whole body FDG-PET/CT scans. At the static level, this setting can be modeled as a clustering problem on the set of 3D connected components of the whole body PET tumoral segmentation mask carried out by nuclear medicine physicians. At the dynamic level, and ad-hoc algorithm is proposed in order to quantify the cancer spread time evolution which, when combined with other existing indicators, gives rise to the metabolic tumor volume-aggressiveness-spread time evolution chart, a novel tool that we claim that would prove useful in nuclear medicine and oncological clinical or research scenarios. Good performance results of the proposed methodologies both at the clinical and technological level are shown using a dataset of 48 segmented whole body FDG-PET/CT scans.  
  Address  
  Corporate Author Thesis  
  Publisher Place of Publication Editor  
  Language Summary Language Original Title  
  Series Editor Series Title Abbreviated Series Title  
  Series Volume Series Issue Edition  
  ISSN (up) ISBN Medium  
  Area Expedition Conference  
  Notes HuPBA;MILAB Approved no  
  Call Number Admin @ si @ SDE2014b Serial 2548  
Permanent link to this record
 

 
Author Frederic Sampedro; Sergio Escalera; Anna Puig edit  doi
openurl 
  Title Iterative Multiclass Multiscale Stacked Sequential Learning: definition and application to medical volume segmentation Type Journal Article
  Year 2014 Publication Pattern Recognition Letters Abbreviated Journal PRL  
  Volume 46 Issue Pages 1-10  
  Keywords Machine learning; Sequential learning; Multi-class problems; Contextual learning; Medical volume segmentation  
  Abstract In this work we present the iterative multi-class multi-scale stacked sequential learning framework (IMMSSL), a novel learning scheme that is particularly suited for medical volume segmentation applications. This model exploits the inherent voxel contextual information of the structures of interest in order to improve its segmentation performance results. Without any feature set or learning algorithm prior assumption, the proposed scheme directly seeks to learn the contextual properties of a region from the predicted classifications of previous classifiers within an iterative scheme. Performance results regarding segmentation accuracy in three two-class and multi-class medical volume datasets show a significant improvement with respect to state of the art alternatives. Due to its easiness of implementation and its independence of feature space and learning algorithm, the presented machine learning framework could be taken into consideration as a first choice in complex volume segmentation scenarios.  
  Address  
  Corporate Author Thesis  
  Publisher Place of Publication Editor  
  Language Summary Language Original Title  
  Series Editor Series Title Abbreviated Series Title  
  Series Volume Series Issue Edition  
  ISSN (up) ISBN Medium  
  Area Expedition Conference  
  Notes HuPBA;MILAB Approved no  
  Call Number Admin @ si @ SEP2014 Serial 2550  
Permanent link to this record
 

 
Author Daniel Sanchez; Miguel Angel Bautista; Sergio Escalera edit  doi
openurl 
  Title HuPBA 8k+: Dataset and ECOC-GraphCut based Segmentation of Human Limbs Type Journal Article
  Year 2015 Publication Neurocomputing Abbreviated Journal NEUCOM  
  Volume 150 Issue A Pages 173–188  
  Keywords Human limb segmentation; ECOC; Graph-Cuts  
  Abstract Human multi-limb segmentation in RGB images has attracted a lot of interest in the research community because of the huge amount of possible applications in fields like Human-Computer Interaction, Surveillance, eHealth, or Gaming. Nevertheless, human multi-limb segmentation is a very hard task because of the changes in appearance produced by different points of view, clothing, lighting conditions, occlusions, and number of articulations of the human body. Furthermore, this huge pose variability makes the availability of large annotated datasets difficult. In this paper, we introduce the HuPBA8k+ dataset. The dataset contains more than 8000 labeled frames at pixel precision, including more than 120000 manually labeled samples of 14 different limbs. For completeness, the dataset is also labeled at frame-level with action annotations drawn from an 11 action dictionary which includes both single person actions and person-person interactive actions. Furthermore, we also propose a two-stage approach for the segmentation of human limbs. In a first stage, human limbs are trained using cascades of classifiers to be split in a tree-structure way, which is included in an Error-Correcting Output Codes (ECOC) framework to define a body-like probability map. This map is used to obtain a binary mask of the subject by means of GMM color modelling and GraphCuts theory. In a second stage, we embed a similar tree-structure in an ECOC framework to build a more accurate set of limb-like probability maps within the segmented user mask, that are fed to a multi-label GraphCut procedure to obtain final multi-limb segmentation. The methodology is tested on the novel HuPBA8k+ dataset, showing performance improvements in comparison to state-of-the-art approaches. In addition, a baseline of standard action recognition methods for the 11 actions categories of the novel dataset is also provided.  
  Address  
  Corporate Author Thesis  
  Publisher Place of Publication Editor  
  Language Summary Language Original Title  
  Series Editor Series Title Abbreviated Series Title  
  Series Volume Series Issue Edition  
  ISSN (up) ISBN Medium  
  Area Expedition Conference  
  Notes HuPBA;MILAB Approved no  
  Call Number Admin @ si @ SBE2015 Serial 2552  
Permanent link to this record
Select All    Deselect All
 |   | 
Details

Save Citations:
Export Records: