lzx551402 / gl3d Goto Github PK
View Code? Open in Web Editor NEWGL3D (Geometric Learning with 3D Reconstruction): a large-scale database created for 3D reconstruction and geometry-related learning problems
License: MIT License
GL3D (Geometric Learning with 3D Reconstruction): a large-scale database created for 3D reconstruction and geometry-related learning problems
License: MIT License
Hello,
I have downloaded your gl3d_gt_pairs file for evaluation from the mirror project, and also, I have downloaded the GL3D ground truth datasets of common_track.txt and mesh_overlap.txt.
However, I do not understand how you generate the weight of the gl3d_gt_pairs file. I read your paper and think the process may be different from what you describe in your paper.
For example, the first test dataset is 584ad76bfe3cb463906ce6dc, in this dataset, the 0 4 weight of the gl3d_gt_pairs file is 0.706303523876, the 0 4 weight of common_track.txt is 0.205568, the 0 4 weight of mesh_overlap.txt is 0.36.
Thanks a lot for your help.
All the links of the datasets on onedrive became unaccessable, please help to check
Hello, thanks for making this benchmark and dataset public.
I'm a bit old school here, but do you think it would be possible to make the full res dataset public (4000x3000 images) to benchmark handcrafted methods against GL3D?
thanks!
After extracting the undistorted images (1000x1000), I found there are multiple images missing, e.g. 00000004/undist_images/00000000.jpg, 00000005/undist_images/00000417.jpg and so on. Totally there are 5631 images missing according to my observation.
I tried to download and extract the data twice, with the same issue. Could you check whether it is an error or my problem? Thanks!
bash download_data.sh gl3d_imgs 0 125
and the error is as below--2020-07-07 21:04:36-- http://research.altizure.com/data/gl3d_imgs/gl3d_imgs.tar.000
Resolving research.altizure.com (research.altizure.com)... 13.33.171.3, 13.33.171.36, 13.33.171.49, ...
Connecting to research.altizure.com (research.altizure.com)|13.33.171.3|:80... connected.
HTTP request sent, awaiting response... 301 Moved Permanently
Location: https://research.altizure.com/data/gl3d_imgs/gl3d_imgs.tar.000 [following]
--2020-07-07 21:04:37-- https://research.altizure.com/data/gl3d_imgs/gl3d_imgs.tar.000
Connecting to research.altizure.com (research.altizure.com)|13.33.171.3|:443... connected.
HTTP request sent, awaiting response... 403 Forbidden
2020-07-07 21:04:37 ERROR 403: Forbidden.
<Error>
<Code>AccessDenied</Code>
<Message>Access Denied</Message>
<RequestId>5EF0B01838FCCBDE</RequestId>
<HostId>
GuVoU7z/ICFJN00nsCA4Prn08kR+8Nl8KC9Jzeeceqg8+bOr0cnxpxJ4qLzOpPMEgYWnkMTgHGs=
</HostId>
</Error>
Hi, will you provide the rendered depth for tourism dataset?
It seems the current download provides undist_images and blended_images for 543 PIDs and raw_images for 40 PIDs.
However it seems it do not include geolabel nor depths data.
Please suggest to download this data.
Thank you
Hi,
I use your script to download the data.
gl3d is ok, but the tourism cams is forbidden to download.
Could you check the right to access the data.
Hello,
I assume the geolabel/cameras.txt has camera intrinsic parameters corresponding to the scale of the raw images. I was wondering how the focal length gets scaled down to match the width and height of the undistorted images (whose aspect ratio also varies from the raw images)
Hi! Thanks for your great work!
I successfully downloaded GL3D dataset by the URL link in readme.
However, I failed to download tourism raw images because I keep getting "HTTP ERROR 502'' from the https://research.altizure.com/ link.
I tried several ip addresses and I still can't solve this problem.
Can someone please tell me what to do next or kindly share the whole dataset?
Thank you!
Line 143 in 021999a
You said
Again, use download_data.sh to fetch the above geometric labels or reconstruction results,
but which word should be used to replace 'gl3d_imgs' in this command?
bash download_data.sh gl3d_imgs 0 125
Hello, where is the download link for the tourism_data dataset?
Hi
Thanks a lot for your work and the dataset.
However, when I download the dataset following your instructions:
bash download_data.sh gl3d_imgs 0 125
The error shows:
wget: 无法解析主机地址 “research.altizure.com”
Is there any trouble with this website?
Looking forward to your reply.
Hello,
I have downloaded your gl3d_gt_pairs file for evaluation from the mirror project, and also, I have downloaded the GL3D ground truth datasets of common_track.txt and mesh_overlap.txt.
However, I do not understand how you generate the weight of the gl3d_gt_pairs file. I read your paper and think the process may be different from what you describe in your paper.
For example, the first test dataset is 584ad76bfe3cb463906ce6dc, in this dataset, the 0 4 weight of the gl3d_gt_pairs file is 0.706303523876, the 0 4 weight of common_track.txt is 0.205568, the 0 4 weight of mesh_overlap.txt is 0.36.
Thanks a lot for your help.
According to the way you gave, the dataset cannot be downloaded successfully and can only be downloaded one by one according to one of the answers in the question.
Since the keypoints are detected from distorted images,how to get patches from undistort images ?
Hello,
I am confused about the data format img_kpts/<img_idx>.bin.
If i get a keypoint detect by SIFT,how can i get the values as:
affine = (-0.25711258, 0.52153266, -0.58181703, -0.78229898, -0.38566887, -0.53048758)
What laws are there between affine[0], affine[1], affine[3], affine[4] ?
Why the affine transformation matrix in <img_idx>.bin is not like this:
affine_mat[0, 0] = m_cos
affine_mat[1, 0] = m_sin
affine_mat[2, 0] = ptf[0]
affine_mat[0, 1] = -m_sin
affine_mat[1, 1] = m_cos
affine_mat[2, 1] = ptf[1]
i read the article GeoDesc: Learning Local Descriptors by Integrating Geometry Constraints,but still confused,
Thanks a lot for your help.
Hello, thank you very much for sharing. After downloading the data, I use cat download_data_gl3d_imgs/.tar. | tar -xvf - to extract the file and display errors. How can I solve this problem
data/585ee0632a57cc11d4933608/raw_images/DJI_0494.JPG
data/585ee0632a57cc11d4933608/raw_images/DJI_0482.JPG
tar: 归档文件中异常的 EOF
tar: 归档文件中异常的 EOF
tar: Error is not recoverable: exiting now
Hi, is your camera.txt under world-to-camera coordinate of camera-to-world coordinate?
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.