{"title":"Robust 2D-3D matching for 3D object pose detection","authors":"Tatsuya Kobayashi, H. Kato, A. Yoneyama","doi":"10.1145/2407746.2407757","DOIUrl":"https://doi.org/10.1145/2407746.2407757","url":null,"abstract":"We present a novel 2D-3D matching method, which is applicable to the pose detection of a target 3D object for an AR application. Our method matches 2D keypoints against their corresponding 3D points directly with a classification approach, and implements a matching refinement using geometric restriction. Our method achieves robust matching and handles the problem that the performance of the conventional keyframe-based methods relies on the pose of the target object in reference images due to self-occlusion. In our experiment, our method improved the matching performance of the conventional method by 15 percentage points.","PeriodicalId":201780,"journal":{"name":"SIGGRAPH Asia 2012 Technical Briefs","volume":"68 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124086750","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}
{"title":"SIGGRAPH Asia 2012 Technical Briefs","authors":"Zhengyou Zhang, Zhengguo Li","doi":"10.1145/2407746","DOIUrl":"https://doi.org/10.1145/2407746","url":null,"abstract":"The SIGGRAPH Asia 2012 Technical Briefs program is a new initiative and will be a premier international forum for new research results in graphics, especially at the intersections of graphics with audio, image, video, and human-computer interaction. \u0000 \u0000Leading international experts from over the world will present their results in peer-reviewed research spanning a wide range of research areas including, but not limited to sensor-based human-computer interaction, graphics and human-computer interface, high-dynamic range imaging, high-definition imaging, 3D imaging, image processing for graphics, high-dynamic range and high-definition video, 3D video, 3D audio, mixed reality, augmented reality, virtual reality, and graphics and multimedia. Accepted papers will be published in the ACM Digital Library, and indexed by EI.","PeriodicalId":201780,"journal":{"name":"SIGGRAPH Asia 2012 Technical Briefs","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114933197","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}
{"title":"3D Diff: an interactive approach to mesh differencing and conflict resolution","authors":"J. Dobos, A. Steed","doi":"10.1145/2407746.2407766","DOIUrl":"https://doi.org/10.1145/2407746.2407766","url":null,"abstract":"We present an open source tool, 3D Diff, that supports differencing and merging of 3D models. As modeling software grows in use, and as 3D models get more complex and require input from more users, there is an emerging problem of maintaining large scenes over time. A scene might be concurrently edited by different users and thus there is a need to merge different versions of a 3D model. We frame this problem in a way that is analogous to software merging: firstly, we automatically detect differences in 3D models by noting correspondences and discrepancies between them; secondly we provide an interactive tool to select between such changes in order to effect a merge. To achieve this we introduce the notions of explicit and implicit conflicts in 3D models and develop a prototype implementation to support the differencing and merging processes. We evaluate this tool with users and find the 3D Diff to be an effective way of merging 3D models. We claim that such tools have an important role to play in the maintenance of large models.","PeriodicalId":201780,"journal":{"name":"SIGGRAPH Asia 2012 Technical Briefs","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134285546","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}
M. Ouhyoung, H. Lin, Yi-Ting Wu, Yi-Shan Cheng, Dominik Seifert
{"title":"Unconventional approaches for facial animation and tracking","authors":"M. Ouhyoung, H. Lin, Yi-Ting Wu, Yi-Shan Cheng, Dominik Seifert","doi":"10.1145/2407746.2407770","DOIUrl":"https://doi.org/10.1145/2407746.2407770","url":null,"abstract":"In recent years, depth cameras have become commodity hardware already in many households. That trend allows for an action-driven facial animator system to become commonly available. In this paper, we layout the methods involved in driving high quality CG facial animations by capturing and matching human actions at real-time speeds. We explore two important sub-problems for this issue: (i) head motion tracking and (ii) animation creation for virtual characters. For head motion tracking, we present a 3D template matching framework, achieving real-time performance and yielding results more accurate than current state-of-the-art motion tracking methods. For animation creation, we provide a 3D animation retrieval system that allows artists to easily retrieve and reuse desired animations from an existing database of already rigged facial animations.","PeriodicalId":201780,"journal":{"name":"SIGGRAPH Asia 2012 Technical Briefs","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131221468","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}
{"title":"An interactive editing tool for correcting panoramas","authors":"Junhong Gao, Michael S. Brown","doi":"10.1145/2407746.2407777","DOIUrl":"https://doi.org/10.1145/2407746.2407777","url":null,"abstract":"Constructing panoramas from an image series is a fundamental task for photo-editing. State-of-the-art photo-editing softwares generally provide automatic stitching routines that first perform image warping followed by seam-cutting or blending to hide misalignment artifacts. This latter step to hide registration errors is often crucial for achieving a perceptually seamless panorama. However, when this step fails, there are no customized photo-editing tools to allow the user to efficiently hide visual artifacts. In this paper, we introduce a simple and effective interactive tool for correcting panoramas. In particular, we describe two features: 1) a seam-editing tool that allows the user to modify blending seams in a local manner; and 2) a content-aware snapping tool to help the user better align local image content between overlapping images.","PeriodicalId":201780,"journal":{"name":"SIGGRAPH Asia 2012 Technical Briefs","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128771935","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}
{"title":"Real-time Manga-like depiction based on interpretation of bodily movements by using Kinect","authors":"Daiki Umeda, Tomoaki Moriya, Tokiichiro Takahashi","doi":"10.1145/2407746.2407774","DOIUrl":"https://doi.org/10.1145/2407746.2407774","url":null,"abstract":"In order to enrich visual communications, we propose a real-time system that interprets bodily movements by using Kinect [Kinect 2012]. Spatial and temporal positions of joints of body movements are captured by using Kinect. Based on captured data analysis, intensions of movements are interpreted. Finally, our system can automatically attach various Manga-like effects, such as speed lines, focus lines, or motion blur, to live action movies in real time. In addition, voice recognition function extends universality of our system's communication capabilities.","PeriodicalId":201780,"journal":{"name":"SIGGRAPH Asia 2012 Technical Briefs","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123649689","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}
Jianxiong Xiao, Bryan C. Russell, James Hays, Krista A. Ehinger, A. Oliva, A. Torralba
{"title":"Basic level scene understanding: from labels to structure and beyond","authors":"Jianxiong Xiao, Bryan C. Russell, James Hays, Krista A. Ehinger, A. Oliva, A. Torralba","doi":"10.1145/2407746.2407782","DOIUrl":"https://doi.org/10.1145/2407746.2407782","url":null,"abstract":"An early goal of computer vision was to build a system that could automatically understand a 3D scene just by looking. This requires not only the ability to extract 3D information from image information alone, but also to handle the large variety of different environments that comprise our visual world. This paper summarizes our recent efforts toward these goals. First, we describe the SUN database, which is a collection of annotated images spanning 908 different scene categories. This database allows us to systematically study the space of possible everyday scenes and to establish a benchmark for scene and object recognition. We also explore ways of coping with the variety of viewpoints within these scenes. For this, we have introduced a database of 360° panoramic images for many of the scene categories in the SUN database and have explored viewpoint recognition within the environments. Finally, we describe steps toward a unified 3D parsing of everyday scenes: (i) the ability to localize geometric primitives in images, such as cuboids and cylinders, which often comprise many everyday objects, and (ii) an integrated system to extract the 3D structure of the scene and objects depicted in an image.","PeriodicalId":201780,"journal":{"name":"SIGGRAPH Asia 2012 Technical Briefs","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134045817","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}
{"title":"Stroke-based real-time ink wash painting style rendering for geometric models","authors":"Tianchen Xu, Lijie Yang, E. Wu","doi":"10.1145/2407746.2407765","DOIUrl":"https://doi.org/10.1145/2407746.2407765","url":null,"abstract":"This paper presents a novel approach for real-time ink wash painting style rendering on geometric models. We first create an ink footprint model in particle form, then extract classified strokes from ordinary 3D models by utilizing the existing geometry information, and finally combine the strokes and the ink to generate an immersive painted image. By taking advantage of parallel processing on GPU, consequently, the rendering of geometric models with realistic features of Chinese ink painting can be performed on a consumer-level PC in real-time.","PeriodicalId":201780,"journal":{"name":"SIGGRAPH Asia 2012 Technical Briefs","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133791899","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}
Shaojing Fan, T. Ng, Jonathan S. Herberg, Bryan L. Koenig, Shiqing Xin
{"title":"Real or Fake?: human judgments about photographs and computer-generated images of faces","authors":"Shaojing Fan, T. Ng, Jonathan S. Herberg, Bryan L. Koenig, Shiqing Xin","doi":"10.1145/2407746.2407763","DOIUrl":"https://doi.org/10.1145/2407746.2407763","url":null,"abstract":"While efforts to create realistic images have generated break-throughs in computer graphics modeling, there has been little research to date on factors causing people to perceive images as real versus computer-generated (CG). The motivation of the current experiment is to begin investigating such factors. We showed both computer-graphics experts and laypersons real photos and CG images. Photo and CG images were of three types: original, modified to show only intrinsic reflectance components, and modified to show only intrinsic shading components (grayscale). Participants judged whether each image was a real photo or a CG image. Results showed that visual realism depends not only on image properties, but also on cognitive characteristics of viewers. Shading was especially crucial for visual realism. Color was also important. Experts outperformed laypersons, but their advantage was limited to grayscale images. This research at the interface between human cognition and computer vision is a starting point for investigating the factors underlying visual realism.","PeriodicalId":201780,"journal":{"name":"SIGGRAPH Asia 2012 Technical Briefs","volume":"216 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115728867","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}
{"title":"Latent nonuniform splines for animation approximation","authors":"Tomohiko Mukai","doi":"10.1145/2407746.2407749","DOIUrl":"https://doi.org/10.1145/2407746.2407749","url":null,"abstract":"This paper presents a new method to approximate animation sequences through a nonlinear analysis of the spatiotemporal data. The main idea is to find a spline curve which best approximates a multivariate animation sequence in a reduced subspace. Our method first eliminates data redundancy among multiple animation channels using principal component analysis (PCA). The reduced sequence of latent variables is then approximated by a nonuniform spline with free knots. To solve the highly-nonlinear multimodal problem of the knot optimization, we introduce a stochastic algorithm called covariance matrix adaptation evolution strategy (CMA-ES). Our method optimizes the control points and the free knots using least-square method and CMA-ES, which guarantees the best approximation for arbitrary animation sequences such as mesh animations and motion capture data. Moreover, our method is applicable to practical production pipeline because both PCA-and CMA-based algorithms are computationally stable, efficient, and quasi manual parameter-free. We demonstrate the capability of the proposed method through comparative experiments with a common approximation technique.","PeriodicalId":201780,"journal":{"name":"SIGGRAPH Asia 2012 Technical Briefs","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129509809","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}