3D perception of object shapes from RGB image input is fundamental towards semantic scene understanding, grounding image-based perception in our spatially 3-dimensional real-world environments. To achieve a mapping between image views of objects and 3D shapes, we leverage CAD model priors from existing large-scale databases, and propose a novel approach towards constructing a joint embedding space between 2D images and 3D CAD models in a patch-wise fashion -- establishing correspondences between patches of an image view of an object and patches of CAD geometry. This enables part similarity reasoning for retrieving similar CADs to a new image view without exact matches in the database. Our patch embedding provides more robust CAD retrieval for shape estimation in our end-to-end estimation of CAD model shape and pose for detected objects in a single input image. Experiments on in-the-wild, complex imagery from ScanNet show that our approach is more robust than state of the art in real-world scenarios without any exact CAD matches.
3D Pose Estimation and 3D Model Retrieval for Objects in the Wild
Institute of Computer Graphics and Vision, Graz University of Technology(格拉茨技术大学计算机图形学与视觉研究所);Laboratoire Bordelais de Recherche en Informatique, University of Bordeaux(波尔多大学波尔多计算机科学实验室)
Fraunhofer IOSB(弗劳恩霍夫光电、系统技术及图像处理研究所);Karlsruhe Institute of Technology KIT(卡尔斯鲁厄理工学院)
目前的logo检索研究主要集中在闭集场景。我们认为,标识域太大,这种策略,需要一个开放的方法。为了促进这一方向的研究,收集并向公众发布了一个名为Logos in the Wild的大型徽标数据集。一个典型的开集标志检索应用是,例如,评估广告在体育赛事广播的有效性。给定徽标图像形状的查询样本,任务是在一组图像或视频中找到该徽标的所有进一步出现。目前,常见的标志检索方法是不适合这项任务,因为他们的封闭世界的假设。因此,在这项工作中,提出了一个开集的标志检索方法,它允许搜索以前看不见的标志由一个单一的查询样本。提出了一个具有单独的标识检测和比较的两阶段概念,其中两个模块都基于特定于任务的CNN。如果使用Logos in the Wild数据进行训练,可以观察到显着的性能改进,特别是与最先进的闭集方法相比。
SHREC'22 Track: Sketch-Based 3D Shape Retrieval in the Wild
Nanjing University of Aeronautics and Astronautics(南京航空航天大学);New York University(纽约大学);Beihang University(北京航空航天大学);IMT Nord Europe(北欧国立高等工业技术与矿业学校);Inception Institute of Artificial Intelligence(起源人工智能研究院);New York University Abu Dhabi(纽约大学阿布扎比分校);University of Science, VNUHCM(胡志明市国家大学科学大学);Viet Nam National University, Ho Chi Minh City(胡志明市越南国家大学)
All in One Framework for Multimodal Re-identification in the Wild
National Engineering Research Center for Multimedia Software(国家多媒体软件工程技术研究中心);Institute of Artificial Intelligence(人工智能研究院);School of Computer Science(计算机学院);Hubei Luojia Laboratory(湖北珞珈实验室);Wuhan University(武汉大学);Guangzhou Urban Planning Design Survey Research Institute(广州市城市规划勘测设计研究院)
Deep Lesion Graphs in the Wild: Relationship Learning and Organization of Significant Radiology Image Findings in a Diverse Large-scale Lesion Database
National Institutes of Health Clinical Center(美国国立卫生研究院临床中心);NVIDIA(英伟达)
Aligning Step-by-Step Instructional Diagrams to Video Demonstrations
The Australian National University(澳大利亚国立大学);Mitsubishi Electric Research Labs(三菱电机研究院);Technion Israel Institute of Technology(以色列理工学院);The Australian Institute for Machine Learning(澳大利亚机器学习研究所)
Recognizing Families through Images with Pretrained Encoder
Faculty of Information Technology, University of Science, VNU-HCM(越南国家大学胡志明市分校科学大学信息技术学院)
亲属关系验证和亲属关系检索是计算机视觉中的新兴任务。亲属关系验证的目的是确定两个人脸图像是否来自相关的人,而亲属关系检索是从图像库中检索可能与一个人相关的人脸图像的任务。他们引入了独特的挑战,因为隐藏的关系和功能,进行面部图像之间的固有特性。我们采用3种方法,FaceNet,Siamese VGG-Face以及FaceNet和VGG-Face模型的组合作为特征提取器,在2020年的Recognizing Family in The Wild比赛中获得了第9名的亲属关系验证和第5名的亲属关系检索。然后,我们使用StyleGAN 2作为另一个编码器进行了进一步的实验,结果没有任何改善。
3D-LENS: A 3D Lifting-based Elevated Novel-view Synthesis method for Single-View Aerial-Ground Re-Identification
Université Paris-Saclay, CEA, List(巴黎萨克雷大学、CEA、List);Sorbonne University, CNRS, Institute of Intelligent Systems and Robotics (ISIR)(索邦大学、CNRS、智能系统与机器人研究所)