Abstract:Reconstruction of three-dimensional (3D) scenes is an active research topic in the field of computer vision and 3D display. It's a challenge to model 3D objects rapidly and effectively. A 3D model can be extracted from multiple images. The system only requires a sequence of images taken with cameras without knowing the parameters of camera, which provide flexibility to a high degree. We focus on quickly merging point cloud of the object from depth map sequences. The whole system combines algorithms of different areas in computer vision, such as camera calibration, stereo correspondence, point cloud splicing and surface reconstruction. The procedure of 3D reconstruction is decomposed into a number of successive steps. Firstly, image sequences are received by the camera freely moving around the object. Secondly, the scene depth is obtained by a non-local stereo matching algorithm. The pairwise is realized with the Scale Invariant Feature Transform (SIFT) algorithm. An initial matching is then made for the first two images of the sequence. For the subsequent image that is processed with previous image, the point of interest corresponding to ones in previous images are refined or corrected. The vertical parallax between the images is eliminated. The next step is to calibrate camera, and intrinsic parameters and external parameters of the camera are calculated. Therefore, The relative position and orientation of camera are gotten. A sequence of depth maps are acquired by using a non-local cost aggregation method for stereo matching. Then point cloud sequence is achieved by the scene depths, which consists of point cloud model using the external parameters of camera and the point cloud sequence. The point cloud model is then approximated by a triangular wire-frame mesh to reduce geometric complexity and to tailor the model to the requirements of computer graphics visualization systems. Finally, the texture is mapped onto the wire-frame model, which can also be used for 3D display. According to the experimental results, we can reconstruct a 3D point cloud model more quickly and efficiently than other methods.

Precise 3d reconstruction from a single image

A Simple 3D Reconstruction Method Based on Single Camera

A Flexible Method for 3D Reconstruction of Urban Building

Sketch-based 3-D modeling for piecewise planar objects in single images

An Example-Based Approach to 3D Man-Made Object Reconstruction from Line Drawings

A Method of Rapid Interactive 3D Reconstruction from Sequence Image

Complete 3D Model Reconstruction Using a Depth Sensor

3D reconstruction algorithm from uncalibrated images with high precision

An Interactive Image-based Modeling System

Outdoor Scene 3D Reconstruction from Multiple Point Cloud

A 3D Reconstruction Framework from Image Sequences Based on Point and Line Features

Reconstruction Of 3d Scenes From Sequences Of Images

A Novel 3d Reconstruction Method from A Single Uncalibrated Camera

3D Reconstruction from a Single Still Image Based on Monocular Vision of an Uncalibrated Camera

A Highly Robust Automatic 3D Reconstruction System Based on Integrated Optimization by Point Line Features

Multiple-image-based 3D Reconstruction

Precise Reconstruction of Geometric Primitives in Built Environments

3D Reconstruction of Feature Point on Object Surface from a Single Image

Object Reconstruction by Integrating 3D Discrete Points and Digital Images

Three-Dimensional Object Surface Reconstruction Based on Camera Calibration and SIFT

A Novel Method for 3D Reconstruction on Uncalibrated Images.