-
-
Save BigglesZX/4016539 to your computer and use it in GitHub Desktop.
import os | |
from PIL import Image | |
''' | |
I searched high and low for solutions to the "extract animated GIF frames in Python" | |
problem, and after much trial and error came up with the following solution based | |
on several partial examples around the web (mostly Stack Overflow). | |
There are two pitfalls that aren't often mentioned when dealing with animated GIFs - | |
firstly that some files feature per-frame local palettes while some have one global | |
palette for all frames, and secondly that some GIFs replace the entire image with | |
each new frame ('full' mode in the code below), and some only update a specific | |
region ('partial'). | |
This code deals with both those cases by examining the palette and redraw | |
instructions of each frame. In the latter case this requires a preliminary (usually | |
partial) iteration of the frames before processing, since the redraw mode needs to | |
be consistently applied across all frames. I found a couple of examples of | |
partial-mode GIFs containing the occasional full-frame redraw, which would result | |
in bad renders of those frames if the mode assessment was only done on a | |
single-frame basis. | |
Nov 2012 | |
''' | |
def analyseImage(path): | |
''' | |
Pre-process pass over the image to determine the mode (full or additive). | |
Necessary as assessing single frames isn't reliable. Need to know the mode | |
before processing all frames. | |
''' | |
im = Image.open(path) | |
results = { | |
'size': im.size, | |
'mode': 'full', | |
} | |
try: | |
while True: | |
if im.tile: | |
tile = im.tile[0] | |
update_region = tile[1] | |
update_region_dimensions = update_region[2:] | |
if update_region_dimensions != im.size: | |
results['mode'] = 'partial' | |
break | |
im.seek(im.tell() + 1) | |
except EOFError: | |
pass | |
return results | |
def processImage(path): | |
''' | |
Iterate the GIF, extracting each frame. | |
''' | |
mode = analyseImage(path)['mode'] | |
im = Image.open(path) | |
i = 0 | |
p = im.getpalette() | |
last_frame = im.convert('RGBA') | |
try: | |
while True: | |
print "saving %s (%s) frame %d, %s %s" % (path, mode, i, im.size, im.tile) | |
''' | |
If the GIF uses local colour tables, each frame will have its own palette. | |
If not, we need to apply the global palette to the new frame. | |
''' | |
if not im.getpalette(): | |
im.putpalette(p) | |
new_frame = Image.new('RGBA', im.size) | |
''' | |
Is this file a "partial"-mode GIF where frames update a region of a different size to the entire image? | |
If so, we need to construct the new frame by pasting it on top of the preceding frames. | |
''' | |
if mode == 'partial': | |
new_frame.paste(last_frame) | |
new_frame.paste(im, (0,0), im.convert('RGBA')) | |
new_frame.save('%s-%d.png' % (''.join(os.path.basename(path).split('.')[:-1]), i), 'PNG') | |
i += 1 | |
last_frame = new_frame | |
im.seek(im.tell() + 1) | |
except EOFError: | |
pass | |
def main(): | |
processImage('foo.gif') | |
processImage('bar.gif') | |
if __name__ == "__main__": | |
main() |
@fjania You're most welcome!
Awesome work! I needed this. Minor nitpick though: there's a weird semantic division here between partial-size frames and frames that are full size, but have transparent regions. In the former we'll output visually-opaque frames but in the latter we'll output partially transparent frames. That is, partial-syle GIFs will return what the GIF looks like during that frame, but full-style GIFs will return only that frame's delta.
Example: http://imgur.com/a/xuK0T
IMO, pasting the previous frames should be an option independent from the partial size detection. If the user doesn't choose to paste, we can handle partials by offsetting the frame the right amount on a transparent base layer.
Also, strictly speaking, it looks like tile[1] returns (x1, y1, x2, y2), not (x1, y1, width, height) so analyseImage should check if the tile starts at 0, 0 as well, in case we have a partial that still reaches to the bottom right corner.
update_region = tile[1]
if update_region != (0, 0,) + im.size:
Thank you for the grate work! But I have a very little question to ask:
-) I am currently using your code and it's working great, however for some gif images, the frames are fetched a little bit distorted with this code and I couldn't figure it out yet.
The gif file:
Do you have any idea why this is happening?
Thanks for this!
I made a fork that supplies frames as returns from a generator: https://gist.github.com/almost/d2832d0998ad9dfec2cacef934e7d247
Awesome code, ineffable appreciation
Very useful code! Thanks!
nice, but the second to last frame of a transparent gif bleeds into the last frame
hello! this still not resolve problems with some GIFs, see examples bellow. Partly it is caused by bug within opening GIF in Python PIL library, see python-pillow/Pillow#2893. There is workaround in that issue that resolve problems with singer GIF bellow but will not help with the other GIF.
If you want to implement GIF processing in your application and you dont want just extract/edit few GIFs for yourself, maybe moviepy library is better option. Dont get me wrong, I think that Python PIL is great for image processing, but it seems that processing GIFs by Python PIL is currently buggy.
UPDATE: looks like there are more reported issues with GIF in Python PIL library https://github.com/python-pillow/Pillow/issues?q=is%3Aissue+is%3Aopen+gif+label%3A%22Palette+Issue%22
Thanks @JankesJanco – it's been many years since I went anywhere near this, but I'm glad it's been useful for some folks! I agree that using a library like moviepy is probably going to be more reliable these days.
Hi Biggles,
you did a great job! There are many, many ... many coders searching for this solution.
Best regards
Axel Arnold Bangert - Herzogenrath 05.03.2020
How would you save the changed frames inside the gif as a new animated gif? With this we could operate on frames inside the gif and save it.
Axel
Hi Axel, that's a little outside of the scope of this code I'm afraid – I also haven't touched it in many years so there are probably better solutions out there by now. moviepy is mentioned above, that might help. In any case I hope it was helpful.
Hi Axel – that's great, always nice to hear from other people. Buying extra pasta in London today :). Good luck with your project and stay healthy! James
Due to PIL updates, to continue using this file as is this would be a possible solution
if not im.getpalette() and im.mode in ("L", "LA", "P", "PA"):
im.putpalette(p)
Thank you sir!