XFusion
XFusion is a deep learning-based spatiotemporal fusion method for high-fidelity ultra-high-speed x-ray radiography. It reconstructs high-quality x-ray image sequences by combining the high spatial resolution of a high-speed camera with the high temporal resolution of an ultra-high-speed camera sequence.
Documentation: https://xrayfusion.readthedocs.io/
If you use XFusion, we kindly request that you cite [B46] and [B44].
User Data (Testing Case)
This dataset is provided as an actual user data testing case (dataset3.zip) and includes paired low- and high-resolution
radiography sequences.
rad_ID |
00003 |
|---|---|
Image preview |
|
Downloads |
|
Contents |
Low-res + High-res paired sequences |
Package |
dataset3.zip |
Low-res pixel size |
3.82 µm |
Low-res frame rate |
1,000,000 fps |
Low-res image size |
400×250 pixels |
Low-res # images |
256 |
High-res pixel size |
1.31 µm |
High-res frame rate |
71,428.6 fps |
High-res image size |
1280×800 pixels |
High-res # images |
681 |
TEM Calibrator Data
This dataset provides paired low- and high-resolution sequences collected using a TEM calibrator target.
rad_ID |
00004 |
|---|---|
Image preview |
|
Downloads |
|
Contents |
Low-res + High-res paired sequences |
Low-res pixel size |
3.82 µm |
Low-res frame rate |
1,000,000 fps |
Low-res image size |
400×250 pixels |
Low-res # images |
256 |
High-res pixel size |
1.31 µm |
High-res frame rate |
71,000 fps |
High-res image size |
1280×800 pixels |
High-res # images |
101 |
