Overview

The GigaMVS dataset is composed of 13 ultra-large-scale scenes , with a maximum area of 32007 m2 and an average area of 8667 m2. For each scene, a large number of gigapixel-level images are available, with both wide field-of-view and high spatial resolution. Regarding the evaluation, the dataset provides precisely labelled camera poses , as well as densely laser-scanned ground-truth 3D models, where the noisy point clouds and the transient objects are either carefully polished or eliminated by post-processing. Please refer to our paper on IEEE Xplore for detailed information of pictures and laser-scanned point clouds.

Before downloading the data, you need to read the copyright terms of the dataset, agree to our licence, and submit the application for using the dataset to us by email. Please refer to the Copyright and Download sections below for details.

Copyright

The GigaMVS dataset is available for the academic purpose only. Any researcher who uses the GigaMVS dataset should obey the licence as below:

All of the GigaMVS Dataset (data, annotation and software) are copyright by Smart Imaging Laboratory, Tsinghua-Berkeley Shenzhen Institute and published under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 License. This means that you must attribute the work in the manner specified by the authors, you may not use this work for commercial purposes and if you alter, transform, or build upon this work, you may distribute the resulting work only under the same license.

This dataset is for non-commercial use only. However, if you find yourself or your personal belongings in the data, please contact us, and we will immediately remove the respective images from our servers.

Download

At present, we provide the data of image and video sequence of GigaMVS dataset, as well as the bounding box (and corresponding labels) annotation and group&interaction annotations of the training set.

To download the GigaMVS dataset, please agree on the license and provide the below information via email. We will only take applications from organization email (please DO NOT use the emails from gmail/163/qq). Anyone who uses the GigaMVS dataset should obey the license and send us an email for registration.

Please use the following email template:

-----------------------------------------------------------
To: gigamvs@outlook.com
Subject: Apply for Using GigaMVS Dataset

I am aware of GigaMVS Terms of Use and I confirm to comply with them.

Name:
Organization:
Email:
Tel:
-----------------------------------------------------------

Annotation Description

File structure

The directory of the benchmark data includes 13 zip files, which correspond to the 13 ultra-large-scale scenes:

|--Dataset_Folder

    |--scene1.zip

    |--scene2.zip

    |--scene3.zip

    ....

    |--scene13.zip

Scene Zip Files

Each sceneX.zip compressed package containing images, corresponding calibrated camera parameters and target poses (for novel view synthesis) of one scene. The directory is as follows:

|--sceneX
    |--images
        |--xxx.jpg
        |--xxx.jpg
        ....
        |--xxx.jpg
    |--cams
        |--xxx_cam.txt
        |--xxx_cam.txt
        ....
        |--xxx_cam.txt
    |--render_cams
        |--xxx_cam.txt
        |--xxx_cam.txt
        ....
        |--xxx_cam.txt

Parameter Files

The camera files are organized in the same way as MVSNet. The camera parameter of one image is stored in a cam.txt file. The text file contains the camera extrinsic E = [R|t], intrinsic K and the depth range:

extrinsic
E00 E01 E02 E03
E10 E11 E12 E13
E20 E21 E22 E23
E30 E31 E32 E33

intrinsic
K00 K01 K02
K10 K11 K12
K20 K21 K22

DEPTH_MIN DEPTH_INTERVAL (DEPTH_NUM DEPTH_MAX)

Citation

All technical papers, documents and reports which use the GigaMVS dataset will acknowledge the use of the database and a citation to:

@ARTICLE{zhang2021gigamvs, author={Zhang, Jianing and Zhang, Jinzhi and Mao, Shi and Ji, Mengqi and Wang, Guangyu and Chen, Zequn and Zhang, Tian and Yuan, Xiaoyun and Dai, Qionghai and Fang, Lu}, journal={IEEE Transactions on Pattern Analysis and Machine Intelligence}, title={GigaMVS: A Benchmark for Ultra-large-scale Gigapixel-level 3D Reconstruction}, year={2021}, volume={}, number={}, pages={1-1}, doi={10.1109/TPAMI.2021.3115028}}

Top