Closed AtlantixJJ closed 3 months ago
Hi,
thanks for your interest in the NeRSemble dataset.
We added the color correction for all of the participants now to the dataset.
You can find the download link in the README.md
file (You have to download the README.md
again to see the changes).
The color_correction.zip
file contains affine color transformation matrices (following the procedure described in the NeRSemble paper) for each participant and each of the 16 cameras.
To apply the color correction to an image of person $ID
and camera $SERIAL
, the following code may be used:
affine_color_transform = np.load('color_correction/$ID/$SERIAL.npy')
image = image @ affine_color_transform[:3, :3] + affine_color_transform[np.newaxis, :3, 3]
image = np.clip(image, 0, 1)
Hope this helps
There is only color correction on the 10 sequences used in the paper while for the majority of other data, there is no such file.
I found in your paper that color correction involves a quite complex procedure, including face segmentation and optimal transport solving, which appears to be hard to reproduce.
I wonder if you could release (or point to) the code of doing such preprocessing, or release the correction data for the rest of the data?
Thank you very much.