Upload datasets as Github artifacts
It would be great to create a GitHub release and upload the datasets as artifacts attached to the release. There can be an unlimited number of artifacts per release and each release artifact can have up to 2GB. The dataset could be split into part-based tar/zip/7z archives. I am asking for this feature, as it would simplify the integration of your dataset into the colmap reconstruction benchmark that I am working on here: https://github.com/colmap/colmap/pull/2714
Thank you for the great dataset.
Hi Johannes,
@Linyou Helped create the github release. Now it should be able to download the low-res set by using wget https://github.com/YoYo000/BlendedMVS/releases/download/v1.0.0/BlendedMVS.zip.
Thanks for the suggestion. It would be great that BlendedMVS could be integrated into colmap reconstruction benchmark!
Yao
Great, thank you! I assume this currently only includes BlendedMVS but not the + and ++ extensions?
Hi, thanks for providing the GitHub link! Could you also mirror the + and ++ extension on Github as well? There seems to be issues with the OneDrive, as mentioned in other open issues, e.g. https://github.com/YoYo000/BlendedMVS/issues/36.
Hi @ahojnnes @skoppula,
Thanks for the suggestion. We’ve also noticed the OneDrive issue for the + and ++ extensions, so we’ve added GitHub release links for both.
- BlendedMVS1 is the + version
- BlendedMVS2 is the ++ version
Each is split into 42 parts because GitHub limits individual files to 2GB. You can download them with wget using the script below:
#!/bin/bash
# Base URL
# Change to v1.0.2 for BlendedMVS++
BASE_URL="https://github.com/YoYo000/BlendedMVS/releases/download/v1.0.1"
# Change to BlendedMVS2 for BlendedMVS++
# Download split files BlendedMVS1.z01 to BlendedMVS1.z42
for i in $(seq -w 1 42); do
FILE="BlendedMVS1.z$i"
echo "Downloading $FILE..."
wget -c "${BASE_URL}/${FILE}"
done
# Download the final .zip descriptor
ZIP_FILE="BlendedMVS1.zip"
echo "Downloading $ZIP_FILE..."
wget -c "${BASE_URL}/${ZIP_FILE}"
echo "All files downloaded."
Thanks for the suggestion. We’ve also noticed the OneDrive issue for the + and ++ extensions, so we’ve added GitHub release links for both.
- BlendedMVS1 is the + version
- BlendedMVS2 is the ++ version
Each is split into 42 parts because GitHub limits individual files to 2GB. You can download them with wget using the script below:
#!/bin/bash
Base URL
Change to v1.0.2 for BlendedMVS++
BASE_URL="https://github.com/YoYo000/BlendedMVS/releases/download/v1.0.1"
Change to BlendedMVS2 for BlendedMVS++
Download split files BlendedMVS1.z01 to BlendedMVS1.z42
for i in $(seq -w 1 42); do FILE="BlendedMVS1.z$i" echo "Downloading $FILE..." wget -c "${BASE_URL}/${FILE}" done
Download the final .zip descriptor
ZIP_FILE="BlendedMVS1.zip" echo "Downloading $ZIP_FILE..." wget -c "${BASE_URL}/${ZIP_FILE}"
echo "All files downloaded."
Hi @Linyou , it seems BlendedMVS1.z05 is missing in current release. Could you please fix that?
Hi,
I downloaded all the zip split files, but unzippping it shows file #193941: bad zipfile offset (lseek): 2029027328 and decompression fails.
Does anyone have the same problem?
Hi,
I downloaded all the zip split files, but unzippping it shows
file #193941: bad zipfile offset (lseek): 2029027328and decompression fails.Does anyone have the same problem?
Solved: https://stackoverflow.com/questions/70263625/how-to-unzip-multi-part-zip-file-in-terminal