zzh-tech / ESTRNN

[ECCV2020 Spotlight] Efficient Spatio-Temporal Recurrent Neural Network for Video Deblurring
MIT License
314 stars 39 forks source link

About Dataset #42

Closed jkpark0825 closed 11 months ago

jkpark0825 commented 1 year ago

First of all, thank you for the great work.

In the BSD dataset, the beam splitter divides the light to create two images with different exposure times, but there could be some misalignment between the paired images. Is any additional method employed to align the misaligned pixels in these images?

Thank you in advance.

zzh-tech commented 1 year ago

there could be some misalignment between the paired images

We use a laser to physically make the misalignment less.

Is any additional method employed to align the misaligned pixels in these images?

And post-processing (homography) is applied to ensure that the misalignment is less than 1 pixel.