ddowd97 / python-image-morpher Goto Github PK

Hello, What an interesting show!
I tried with photos of cats and dogs, and the results are as follows：

It seems that the effect is not ideal. Do you have any suggestions？
Thanks

With requirements.txt other people can just pip install -r requirements.txt

pyqt5
scipy
numpy
imageio
mathplotlib

Read more

Hi, first of all, it's an amazing project of yours. I was reading your documentation and you write there that the user should be also able to upload the (x,y) coordinates of the landmark points. I have not found the option in your GUI. Is it already implemented? And more importantly, would it be possible to add a button for exporting the points into a txt file?
Thanks a lot.

Great work!
Works perfectly in the Linux-container on my Chromebook.

Suggestions

• graphical mode indicator/selector [add][move][del]
• "Interim Steps" 10 -- 1000 instead blend 0.1 -- 0.001

Keep up the good work.

my result: https://www.instagram.com/tv/CfAH-pmDOjj/?igshid=YmMyMTA2M2Y=

You must change opencv-python for opencv-python-headless in requirements.txt, as when executing the software it crashes as it uses opencv plugins instead of native PyQT ones.

Hi David,

Many thanks for your great image morphing application!

I created this Python command line program based on your work:

https://github.com/jankovicsandras/autoimagemorph

The main difference (beside no GUI) is that autoimagemorph selects the points automatically, using cv2.goodFeaturesToTrack() .

I noticed some image artifacts (bright pixels probably at Delaunay triangle edges) and found a simple method to remove them: upscale the image (for example 4x width and height), do the processing, downscale it to original size again. I call this subpixel processing in autoimagemorph , and the obvious downside is much more time and memory required (approx. 16x slower using -subpixel 4) . Still, it might be a good idea to provide this function optionally for better quality, as it's very easy to implement.

Keep up the good work! :)

The GUI is great and I still can't stop admiring it. I was wondering -- would it be possible to add a functionality that a point can be selected after it's placed and then with a drag moved to another location? Because now if one makes a mistake or is not precise enough, the only fix is to delete the point and then place it again which is a bit annoying. And even more if one realises that this mistake has been done a few moves back :).

And especially with the automatic feature recognition mentioned in #2, that would be a killer ;). I have some code related to automatic face landmark detection which I can share, it's not working great by itself but if the user could then tweak the not fitting parts, it would save the user a lot of time.

I've been looking a bit at the code now, and I don't seem to be able to figure out how I could output the final image as an image file. It would be cool to be able to output all the different frames from a full blend into a gif for example. I don't know if you could point me in a direction where I could figure it out self, or if it maybe could be easy to implement for you, but I would really apreciate it

The .gif exported has (sometimes) compression artifacts. How can a blend created in the Python Image Morpher be exported as an image sequence?
Converting the .gif to an image sequence does not help as the compression artifacts still remain.
Thank You!