CASIA OpenIR  > 毕业生  > 博士学位论文
基于GPU的大场景三维重建和物体建模
其他题名GPU Based Large-Scene 3D Reconstruction and Object Modeling
刘鑫
学位类型工学博士
导师胡占义
2012-10-23
学位授予单位中国科学院大学
学位授予地点中国科学院自动化研究所
学位专业模式识别与智能系统
关键词三维场景重建 物体建模 基于gpu的快速算法实现 特征提取与匹配 捆绑调整 计算机集群 3d Scene Reconstruction Object Modeling Gpu Based Fast Implementation Feature Extraction And Matching Bundle Adjustment Cluster
摘要近年来,随着图形处理器(Graphic Processing Unit,GPU)通用计算框架的普及,GPU被越来越多地用于计算机视觉领域的快速算法实现。本文旨在探索基于GPU的三维快速重建方法,包括特征提取、特征匹配和捆绑调整等。本文的主要工作有: (1) 基于GPU的特征点提取与匹配。针对高分辨率图像, 实现了一种基于GPU的SIFT(Scale-invariant feature transform, SIFT)特征点提取和匹配算法(KD树搜索),并与对应的CPU实现进行了系统比较。比较实验使用Oxford的经典图像数据库和我们自采集的高分辨率图像数据库。实验结果表明,基于GPU的实现和CPU的实现能提取出95%左右的相似特征点。对于视角变换、尺度变换、旋转、模糊和光照变换等各种图像变换,两种算法实现的特征点重复率和匹配率相当,但GPU实现可获得高达13x(特征提取)和43x(特征匹配)的加速比。上述结果表明,基于GPU的实现与基于CPU的实现一样可靠,在具体应用中完全可以安全使用。 (2) 基于GPU的并行捆绑调整算法。为了有效地解决每幅图像中存在大量特征点的捆绑调整问题–MPBA(Massive-Points Bundle Adjustment, MPBA)问题,提出了一种基于多GPU和多核CPU的并行捆绑调整算法。算法不需要消耗大量显存,有利于在大规模捆绑调整问题上使用。使用高分辨率图像数据库,我们生成了若干MPBA问题,并对几种捆绑调整算法进行了系统评估。与经典稀疏捆绑调整算法(Sparse Bundle Adjustment, SBA)相比,本文算法在保持算法精度的前提下获得了计算效率的显著提升。针对MPBA问题,本文还研究了几种捆绑调整的加速策略。 (3) 基于集群的分布式捆绑调整算法。针对大规模的MPBA问题,提出了一种分布式的捆绑调整算法。为了分解大规模MPBA问题,提出了一种基于三维点集划分的分解方法。该分解方法不依赖于输入参数的内在联系,因而分解结果与具体捆绑调整问题无关。针对不同的集群配置,提出了两种分布式的线性方程求解方法。为了测试本文方法的有效性,将算法映射到两个集群上,一个集群有5台计算机,另一个集群有3台计算机,其中每台机器都配置一块GPU。与单机版经典捆绑调整算法SBA相比,本文算法在保持算法精度的前提下获得了计算效率的显著提升,且本文算法的单机内存消耗显著减少,具有良好的可扩展性。 (4) 基于GPU和Kinect的快速物体建模及精度评估。本文提出了一种基于GPU和多个Kinect传感器的快速物体建模方法,以及对该方法的精度评估。本文方法主要分为两步:一是系统的初始标定,二是全自动的物体建模。对于系统初始标定,提出了一种简单易用的粗标定方法。对于物体建模,提出一种全自动的快速物体建模方法。针对Loop-Closure问题,提出了一种全局的点云配准方法。针对模型融合问题,提出一种对多个点云进行快速融合的方法。本文方法鲁棒性高,在出现点云配准错误时仍然能够稳定地得到较理想的重建模型。我们将本文的物体重建结果与三维激光扫描仪获得的真实物体模型进行了系统比较。实验结果表明,基于Kinect重建的物体模型不仅具有较好的视觉效果,而且其平均残留误差与激光扫描仪重建的物体模型相似(在0.763mm-2.473mm...
其他摘要With the popularity and easy access of high resolution images in recent years,many computer vision problems become computationally prohibitive, and GPU based acceleration is considered the most effective way to tackle the problem. This thesis is focused on GPU based scene reconstruction and object modeling, in particular, the involved key steps, such as feature detection, matching, and final optimization, and the main contributions include: (1) The image feature detection and the feature matching. We take two widely used algorithms as representative examples to evaluate the performance of the GPU implementation, that is, the Scale-invariant feature transform (SIFT) for feature detection and the KD-Tree traversal for feature matching. We present the design, implementation, and evaluation of these two algorithms on GPU for high-resolution image datasets. Comparisons are performed on the implementation on CPU and GPU using classical Oxford’s database as well as our own high-resolution image datasets. Our results show that around 95% of the extracted features are nearly the same, and the repeatability score and matching score are similar under the GPU and CPU implementations under various image changes, such as viewpoint change, scale change, rotation, bluring and illumination variation. The runtime speedup of the GPU implementation for the SIFT detection is about 13x faster than their CPU counterpart, and that for the KD-Tree traversal is about 43x. In sum, our results show that the GPU implementations for feature detection and matching are as good as the CPU ones, and can be safely used in real applications. (2) GPU Based Bundle Adjustment(BA). BA is a crucial step in 3D reconstruction but time consuming. In this thesis, a special class of BA problems are investigated, where the reconstructed 3D points are much numerous than the camera parameters, namely Massive-Points BA (MPBA) problems. We present the design and implementation of a new bundle adjustment algorithm for efficiently solving the MPBA problems. The hardware parallelism, the multi-core CPUs as well as GPUs, is explored. By careful memory-usage design, the graphic-memory limitation is effectively alleviated. Several modern acceleration strategies for bundle adjustment are explored and compared. By using several high-resolution image datasets, we generate a variety of MPBA problems, with which the performance of five bundle adjustment algorithms are evaluated. The experimental results show th...
馆藏号XWLW1823
其他标识符200918014628036
语种中文
文献类型学位论文
条目标识符http://ir.ia.ac.cn/handle/173211/6482
专题毕业生_博士学位论文
推荐引用方式
GB/T 7714
刘鑫. 基于GPU的大场景三维重建和物体建模[D]. 中国科学院自动化研究所. 中国科学院大学,2012.
条目包含的文件
文件名称/大小 文献类型 版本类型 开放类型 使用许可
CASIA_20091801462803(4722KB) 暂不开放CC BY-NC-SA请求全文
个性服务
推荐该条目
保存到收藏夹
查看访问统计
导出为Endnote文件
谷歌学术
谷歌学术中相似的文章
[刘鑫]的文章
百度学术
百度学术中相似的文章
[刘鑫]的文章
必应学术
必应学术中相似的文章
[刘鑫]的文章
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。