2018 31st SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)最新文献

A Practical Review on Medical Image Registration: From Rigid to Deep Learning Based Approaches 医学图像配准的实践综述:从僵化到基于深度学习的方法

2018 31st SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) Pub Date : 2018-10-01 DOI: 10.1109/SIBGRAPI.2018.00066

Natan Andrade, F. Faria, F. Cappabianco

引用次数: 16

Graph Spectral Filtering for Network Simplification 用于网络简化的图谱滤波

2018 31st SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) Pub Date : 2018-10-01 DOI: 10.1109/SIBGRAPI.2018.00051

Markus Diego Dias, Fabiano Petronetto, Paola Valdivia, L. G. Nonato

引用次数: 1

A Method for Opinion Classification in Video Combining Facial Expressions and Gestures 一种结合表情和手势的视频意见分类方法

2018 31st SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) Pub Date : 2018-10-01 DOI: 10.1109/SIBGRAPI.2018.00011

Airton Gaio Junior, E. Santos

{"title":"A Method for Opinion Classification in Video Combining Facial Expressions and Gestures","authors":"Airton Gaio Junior, E. Santos","doi":"10.1109/SIBGRAPI.2018.00011","DOIUrl":"https://doi.org/10.1109/SIBGRAPI.2018.00011","url":null,"abstract":"Most of the researches dealing with video-based opinion recognition problems employ the combination of data from three different sources: video, audio and text. As a consequence, they are solutions based on complex and language-dependent models. Besides such complexity, it may be observed that these current solutions attain low performance in practical applications. Focusing on overcoming these drawbacks, this work presents a method for opinion classification that uses only video as data source, more precisely, facial expression and body gesture information are extracted from online videos and combined to lead to higher classification rates. The proposed method uses feature encoding strategies to improve data representation and to facilitate the classification task in order to predict user's opinion with high accuracy and independently of the language used in videos. Experiments were carried out using three public databases and three baselines to test the proposed method. The results of these experiments show that, even performing only visual analysis of the videos, the proposed method achieves 16% higher accuracy and precision rates, when compared to baselines that analyze visual, audio and textual data video. Moreover, it is showed that the proposed method may identify emotions in videos whose language is other than the language used for training.","PeriodicalId":208985,"journal":{"name":"2018 31st SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130166289","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

3D Medical Objects Retrieval Approach Using SPHARMs Descriptor and Network Flow as Similarity Measure 基于SPHARMs描述符和网络流的三维医学对象检索方法

2018 31st SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) Pub Date : 2018-10-01 DOI: 10.1109/SIBGRAPI.2018.00049

L. Bergamasco, K. Lima, C. Rochitte, Fátima L. S. Nunes

{"title":"3D Medical Objects Retrieval Approach Using SPHARMs Descriptor and Network Flow as Similarity Measure","authors":"L. Bergamasco, K. Lima, C. Rochitte, Fátima L. S. Nunes","doi":"10.1109/SIBGRAPI.2018.00049","DOIUrl":"https://doi.org/10.1109/SIBGRAPI.2018.00049","url":null,"abstract":"The data processing to obtain useful information is a trending topic in the computing knowledge domain since we have observed a high demand arising from society for efficient techniques to perform this activity. Spherical Harmonics (SPHARMs) have been widely used in the three-dimensional (3D) object processing domain. Harmonic coefficients generated by this mathematical theory are considered a robust source of information about 3D objects. In parallel, Ford-Fulkerson is a classical method in graph theory that solves network flows problems. In this work we demonstrate the potential of using SPHARMs along with the Ford-Fulkerson method, respectively as descriptor and similarity measure. This article also shows how we adapted the later to transform it into a similarity measure. Our approach has been validated by a 3D medical dataset composed by 3D left ventricle surfaces, some of them presenting Congestive Heart Failure (CHF). The results indicated an average precision of 90%. In addition, the execution time was 65% lower than a descriptor previously tested. With the results obtained we can conclude that our approach, mainly the Ford-Fulkerson adaptation proposed, has a great potential to retrieve 3D medical objects.","PeriodicalId":208985,"journal":{"name":"2018 31st SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134230406","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 3

Image-Based Visualization of Classifier Decision Boundaries 基于图像的分类器决策边界可视化

2018 31st SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) Pub Date : 2018-10-01 DOI: 10.1109/SIBGRAPI.2018.00052

F. C. M. Rodrigues, R. Hirata, A. Telea

引用次数: 13

Extracting Visual Encodings from Map Chart Images with Color-Encoded Scalar Values 从带有颜色编码标量值的地图图表图像中提取视觉编码

2018 31st SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) Pub Date : 2018-10-01 DOI: 10.1109/SIBGRAPI.2018.00025

Angela Mayhua, Erick Gomez Nieto, Jeffrey Heer, Jorge Poco

引用次数: 6

Single-Shot Person Re-Identification Combining Similarity Metrics and Support Vectors 结合相似性度量和支持向量的单镜头人物再识别

2018 31st SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) Pub Date : 2018-10-01 DOI: 10.1109/SIBGRAPI.2018.00039

Anderson Luis Cavalcanti Sales, R. H. Vareto, W. R. Schwartz, Guillermo Cámara Chávez

引用次数: 1

Inverse Projection of Vector Fields 向量场的逆投影

2018 31st SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) Pub Date : 2018-10-01 DOI: 10.1109/SIBGRAPI.2018.00050

Paula Ceccon Ribeiro, H. Lopes

{"title":"Inverse Projection of Vector Fields","authors":"Paula Ceccon Ribeiro, H. Lopes","doi":"10.1109/SIBGRAPI.2018.00050","DOIUrl":"https://doi.org/10.1109/SIBGRAPI.2018.00050","url":null,"abstract":"Vector fields play an essential role in a large range of scientific applications. They are commonly generated through computer simulations. Such simulations may be a costly process since they usually require an intensive computational time. When researchers want to quantify the uncertainty in such kind of applications, usually an ensemble of vector fields realizations are generated, making the process much more expensive. The main contribution of this paper is to present a new method, based on the inverse projection technique, to quickly and consistently generate 2D vector fields similar to the ones in the ensemble, which after an evaluation of a specialist could enlarge the ensemble in order to better represent the uncertainty. Through the Helmholtz-Hodge Decomposition, we obtain the divergence-free, rotational-free and harmonic components of a vector field. With those components and the original ensemble in hand, it is possible to derive new realizations from their projections into a 2-dimensional space. To do so, we propose the use of an inverse projection technique individually in each component projected space. Results are obtained in real-time, through an interactive interface. A set of multi-method wind forecast realizations were used to demonstrate the results obtained with this approach.","PeriodicalId":208985,"journal":{"name":"2018 31st SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116888001","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1

SIBGRAPI 2018 Foreword

2018 31st SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) Pub Date : 2018-10-01 DOI: 10.1109/sibgrapi.2018.00005

引用次数: 0

Unsupervised Representation Learning Using Convolutional and Stacked Auto-Encoders: A Domain and Cross-Domain Feature Space Analysis 使用卷积和堆叠自编码器的无监督表示学习:域和跨域特征空间分析

2018 31st SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) Pub Date : 2018-10-01 DOI: 10.1109/SIBGRAPI.2018.00063

G. B. Cavallari, Leo Sampaio Ferraz Ribeiro, M. Ponti

{"title":"Unsupervised Representation Learning Using Convolutional and Stacked Auto-Encoders: A Domain and Cross-Domain Feature Space Analysis","authors":"G. B. Cavallari, Leo Sampaio Ferraz Ribeiro, M. Ponti","doi":"10.1109/SIBGRAPI.2018.00063","DOIUrl":"https://doi.org/10.1109/SIBGRAPI.2018.00063","url":null,"abstract":"A feature learning task involves training models that are capable of inferring good representations (transformations of the original space) from input data alone. When working with limited or unlabelled data, and also when multiple visual domains are considered, methods that rely on large annotated datasets, such as Convolutional Neural Networks (CNNs), cannot be employed. In this paper we investigate different auto-encoder (AE) architectures, which require no labels, and explore training strategies to learn representations from images. The models are evaluated considering both the reconstruction error of the images and the feature spaces in terms of their discriminative power. We study the role of dense and convolutional layers on the results, as well as the depth and capacity of the networks, since those are shown to affect both the dimensionality reduction and the capability of generalising for different visual domains. Classification results with AE features were as discriminative as pre-trained CNN features. Our findings can be used as guidelines for the design of unsupervised representation learning methods within and across domains.","PeriodicalId":208985,"journal":{"name":"2018 31st SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125860763","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 23