智能手机三维重建图像与参数数据集

该数据为基于智能手机拍摄的多场景目标三维重建基础数据集合，涵盖静态物体（如雕像、工业零件、建筑构件）、中等复杂度场景（如室内摆件组合、户外小型景观）等 15 + 类目标的多角度拍摄图像及配套参数。通过标准化拍摄流程采集图像，同步记录相机内参（焦距、主点坐标、畸变系数）、拍摄参数（角度间隔、距离、光照强度）及特征点匹配结果，为智能手机三维重建算法优化、相机标定模型训练、特征点提取与匹配技术改进提供数据支撑，可直接用于 SFM 稀疏重建、CMVS 图像聚簇、PMVS 密集匹配等流程的验证与优化。

This dataset is a foundational collection for 3D reconstruction of multi-scene targets captured using smartphones. Covering over 15 categories of targets including static objects (e.g., statues, industrial parts, building components) and moderately complex scenes (e.g., indoor ornament combinations, outdoor small landscapes), it provides multi-angle captured images and supporting parameters. Images were acquired through a standardized shooting protocol, with synchronous recording of camera intrinsic parameters (focal length, principal point coordinates, distortion coefficients), shooting parameters (angle interval, shooting distance, illumination intensity) and feature point matching results. This dataset offers data support for the optimization of smartphone-based 3D reconstruction algorithms, training of camera calibration models, and the improvement of feature point extraction and matching techniques. Moreover, it can be directly used for the verification and optimization of workflows such as SFM sparse reconstruction, CMVS image clustering, and PMVS dense matching.

搜集汇总

背景与挑战

背景概述

该数据集是基于智能手机拍摄的多场景三维重建基础数据集合，包含静态物体和中等复杂度场景的15+类目标的多角度图像及配套相机参数，用于三维重建算法优化和计算机视觉技术研发。数据通过标准化流程采集，记录内参、拍摄参数和特征点匹配结果，支持SFM稀疏重建等流程验证。