F-G Fernandez

Alright, we almost got everything to close this issue: - [x] Check the expected differences in interpolation between cv2, TF & PyTorch - [x] Identify how to bring the difference...

Thanks! So, to even everything out, we would only need to enforce anti-aliasing on TF resize for now? If so, we could: - start doing this in future trainings -...

Let's harmonize this for the next training cycle along with the rotation in 0.5.0 :+1:

We won't have time to harmonize & retrain all models with this before the next release, so I'm staging it for 0.6.0 :+1:

My bad, I'll jump back into this this week

Hi @fharper, Thanks for reporting this! To make sure I get the root problem correctly here, can you confirm that you are able to install DocTR outside of Docker with...

Let's first address #480, I'm rather confident that if it gets fixed, then resolving this issue will be casual :)

Hey there :wave: Actually we have tackled this internally a few weeks back and it will be integrated into docTR soon :smile: But this solution involves a DL model for...

That's a good idea indeed, if you could have a runnable Colab notebook so that we can compare this :+1: (not opening a PR, just sharing it here)

Thanks a lot, it looks promising for single page docs! To make sure this matches the same specs, could you illustrate a situation where there are several pages on the...