Building 3D Shapes from Parts
نویسندگان
چکیده
The research described in this thesis has focused on the composition of novel 3D shapes (e.g., depth maps and triangulated meshes) by reusing parts of existing shapes. Specifically, two methods were designed [28, 27]. The first is a novel tool for easy composition of existing 3D models (i.e., triangulated meshes) into novel models [28]. This tool was designed with the intention of providing a means for rapid modelling by 3D graphics artists. This tool was further shown to have an interface which is simple enough to be automated, to a large extent, for various modelling tasks beyond composition. These tasks include repairing 3D models and filling-in parts of their surfaces. The simplicity of its interface has consequently prompted research into ways of fully automating shape composition from parts. In particular, the second outcome of this research is a method which automatically fuses depth patches from example objects, to produce a depth estimate for a novel object viewed in a single image [27]. A by-product of this reconstruction method, was the development of a system for automatic colorization of depth maps [26], which shares much of its design with the reconstruction procedure. This thesis describes these three methods, and presents related, ongoing and planned research.
منابع مشابه
Simulation of Smoke Emission from Fires in High-Rise Buildings Using the 3D Model Generated from 2-Dimensional Cadastral Data
Having a 3-Dimensional model of high-rise buildings can be used in disaster management such as fire cases to reduce casualties. The fundamental dilemma in 3D building modeling is the unavailability of suitable data sources. However, available cadastral 2D maps could be used as low-cost and attainable resources for 3D building modeling. Smoke will be a great threat to people's health during a f...
متن کاملIntegrated-approach-based Automatic Building Extraction
This paper presents an approach to automatic building extraction from large-scale aerial images. It is based on the integration of multiple information, knowledge and appropriate methods, and hierarchically consists of three major parts: building detection, building segment extraction, 3D feature matching and building modelling. Buildings are detected by segmenting DSMs using grey-scale mathema...
متن کاملGeneralisation of 3d Building Models by Cell Decomposition and Primitive Instancing
The article presents a cartographic generalisation approach for 3D building models with regard to the thematic visualisation of urban landscapes. Based on our earlier work to utilise approximating planes for generating simplified cell decompositions of the input objects, a new extension is introduced that guarantees well-formed roof structures. This is accomplished by first creating a simplifie...
متن کاملUnsupervised Co-segmentation of 3D Shapes via Functional Maps
We present an unsupervised method for co-segmentation of a set of 3D shapes from the same class with the aim of segmenting the input shapes into consistent semantic parts and establishing their correspondence across the set. Starting from meaningful pre-segmentation of all given shapes individually, we construct the correspondence between same candidate parts and obtain the labels via functiona...
متن کاملIntegration of Building Knowledge into Binary Space Partitioning for the Reconstruction of Regularized Building Models
Recent approaches for the automatic reconstruction of 3D building models from airborne point cloud data integrate prior knowledge of roof shapes with the intention to improve the regularization of the resulting models without lessening the flexibility to generate all real-world occurring roof shapes. In this paper, we present a method to integrate building knowledge into the data-driven approac...
متن کامل