mirror of
https://github.com/CCExtractor/ccextractor.git
synced 2026-02-04 05:44:53 +00:00
[BUG] GUI, Burned-in Subtitle Extraction not working #343
Reference in New Issue
Block a user
Delete Branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @bioluminesceme on GitHub (Nov 19, 2017).
CCExtractor version (using the --version parameter preferably) : 0,85
In raising this issue, I confirm the following (please check boxes, eg [X]):
My familiarity with the project is as follows (check one, eg [X]):
Necessary information
What platform did you use?
Windows
What where the used arguments?
C:\Program Files (x86)\CCExtractor\ccextractorwin.exe --gui_mode_reports -in=mp4 -autoprogram -out=srt -bom -latin1 -hardsubx -subcolor white -conf_thresh 60 [+input files]Without OCR in the Input files tab, gives the error Parameter -hardsubx not understood.
C:\Program Files (x86)\CCExtractor\ccextractorwinfull.exe --gui_mode_reports -in=mp4 -autoprogram -out=srt -bom -latin1 -hardsubx -subcolor white -conf_thresh 60 [+input files]With OCR in the Input files tab , it complains about not having enough memory to initialize Tesseract.
I can run Tesseract from the command line using Python on this machine without any issues. Win10, 3.4GHz, 8GB Ram.
Video links
I have a Dropbox link I can share privately.
@uditjuneja commented on GitHub (Feb 14, 2018):
I think the problem can be solved by using destroy() method.
As it can free memory used by object.
@sjtsnix commented on GitHub (Feb 25, 2018):
I am new to CCExtractor (attempted to use it as a part of MCEBuddy on a WTV file that resulted in Garbled random text), so I find my way here, which lead me to install the Windows GUI version of CCExtractor, on my first attempt, I am seeing this same error:
"Not Enough memory to initialize Tesseract!"
This is in the GUI, on a WTV file, Windows 10, 16 GB RAM,
I changed a few settings from the default initially. then after this error, changed back to defaults on the "About & Save" TAB.
I'll be opening separate issue for the garbled text in MCEBuddy
Command line per GUI for modified settings:
C:\Program Files (x86)\CCExtractor\ccextractorwinfull.exe --gui_mode_reports -in=wtv -autoprogram -out=srt -bom -cf C:\Users\sj\Desktop\CCExtractor_Elementary_Stream.txt -goppts -latin1 --endcreditstext "Generated by CCExtractor\nhttp://www.ccextractor.org" --endcreditsforatleast 6 --endcreditsforatmost 3 -hardsubx -subcolor white -conf_thresh 60 [+input files]
@rudolphos commented on GitHub (Apr 11, 2018):
I'm having the same problem.
ccextractorwinfull.exe --gui_mode_reports -autoprogram -out=srt -bom -latin1 --nofontcolor -hardsubx -subcolor white -conf_thresh 60 [+input files]16 GB not enough?...
I also tried on WSL.
But hardsub flag is not even recognized.
~/ccextractor/linux$ ./ccextractor -autoprogram -out=srt -bom -latin1 --nofontcolor -hardsubx -subcolor white -conf_thresh 60 testfile.mkv Error: Error: Parameter -hardsubx not understood. CCExtractor 0.87, Carlos Fernandez Sanz, Volker Quetschke. Teletext portions taken from Petr Kutalek's telxccTo build the program with hardsubx support, from the Linux directory run:- ./configure --enable-hardsubx make ENABLE_HARDSUBX=yesThis is weird, but it doesn't work, many things from the website doesn't work as well, like this tutorial from 2016 - https://abhinavshukla95.wordpress.com/2016/08/18/google-summer-of-code-work-product-submission/
:~/ccextractor/linux$ ./configure --enable-hardsubx -bash: ./configure: No such file or directory~/ccextractor/linux$ ls -a. autogen.sh builddebug build-static.sh configure.ac Makefile.am pre-build.sh.. build build_hardsubx cleanup description-pak module_generator ubuntu@computer:~/ccextractor/linux$ ./configure.ac --enable-hardsubx -bash: ./configure.ac: Permission denied@rudolphos commented on GitHub (Apr 12, 2018):
I compiled it, have all the necessary components, but hardsubx still not recognized.
@cfsmp3 commented on GitHub (Apr 12, 2018):
It's not actually compiling anything:
~/ccextractor/linux$ make ENABLE_HARDSUBX=yes
make: Nothing to be done for 'all'.
On Wed, Apr 11, 2018 at 4:04 PM, rudolphos notifications@github.com wrote:
@rudolphos commented on GitHub (Apr 12, 2018):
I fixed it. But there is another problem:
Can it detect subtitles properly like these by using OCR / burned-in sub detection?
@AntonOfTheWoods commented on GitHub (Nov 17, 2018):
Does anyone have any more info on this? @rudolphos can you give some details on "I fixed it" please? Thanks!
@RobJacobson commented on GitHub (Dec 12, 2018):
Ditto to Anton's request. I'm having exactly the same problem in both Windows and Ubuntu. @rudolphos, could you please explain how you fixed this?
@AntonOfTheWoods commented on GitHub (Dec 12, 2018):
@RobJacobson , I also "fixed" it - the problem is that it seems (at least at the time) Tesseract 4 is not properly supported yet. I was able to get it compiled and extracting subs using the latest Tesseract 3 installed from source. The extraction was not particularly accurate but I will definitely get back to trying to tweak that soon (hopefully).
@RobJacobson commented on GitHub (Jan 12, 2019):
Thanks for the info! Really appreciate it.
@anonynamja commented on GitHub (Jan 16, 2019):
@cfsmp3 @RobJacobson @AntonOfTheWoods
On ubuntu 18.04. I first followed the instructions here: https://github.com/CCExtractor/ccextractor/blob/master/docs/COMPILATION.MD)
Skipping lines with tesseract-ocr and tesseract-ocr-dev since above commenters explained that Tesseract 4 is unsupported, so I used Tesseract 3.05.02 from here: https://github.com/tesseract-ocr/tesseract/archive/3.05.02.tar.gz
Then I followed the instructions here as a model: https://github.com/CCExtractor/ccextractor/blob/master/docs/OCR.md
which produced "Error: avcodec library not found."
How should I proceed?
@atulpatildbz commented on GitHub (Jun 22, 2019):
I get the same error. Couldn't find a solution yet.
@cfsmp3 commented on GitHub (Jan 25, 2020):
@atulpatildbz @anonynamja @RobJacobson Is this still a problem in the current master?
@cfsmp3 commented on GitHub (Mar 22, 2023):
Closing due to no answer