In a continued attempted to find meaning in Prisoner106 I am continuing to analyse the first reel of the surveillance footage by various means.
I had previously downloaded Videogrep a tool for making supercut videos from video and subtitle (.srt) files.
First I had to find a subtitle file for the video. As usual with these things google is helpful. Adding the date of the video, 1967 and eng to the search helped. I could not find one for the dvds I had acquired, but got one for a cd set.
The first run through of the tool
python videogrep.py --input ~/Desktop/the\ prisoner/the-prisoner-ep-1.m4v --search 'six'
Threw a bunch of errors that I could not understand. I headed over to the other confusing village that is github to look for clues. I did a bit of googling too. Eventually I decided just to download a fork of videogrep,habi/videogrep and see what happens. As I had all the dependencies installed I just ran the new version. This did give me a video, but confusingly nothing matched, the timings were out.
It looks like the STR file I had did not match the DVD. I’ve had this problem in the past and usually failed to fix it. But I gave Subtitle ReSync Basic from squixz.com a try and seemed to match things up.
Back to videogrep which ran through its ‘stuff’ and spat out a video, it did seem a bit short so I looked at the output:
[+] Search term 'six' was found in 3 places.
[+] Creating clips.
547.155 to 549.109: Woman Is your number six?
1429.498 to 1432.741: Please go to number six - adjustment is needed.
2195.285 to 2198.724: - Yours is six. - I'm not a number. I'm a person.
Opening the srt file in a text editor I just replaced all the occurrences of Six with six, I should have used the regEx for six or Six or case insensitive, which I’ve done before but I’ve forgotten how and my Internet connection is currently very slow.
I figured out the case for the second supercut:
python videogrep.py --input /Users/john/Desktop/the\ prisoner\ rips/the-prisoner-ep-1.srt --search 'number|Number' -p 1000
the -p flag pads the segments (in miliseconds)
I don’t think my SRT files is properly matched with the video, hopefully I’ll be able to find or make a better match from some subsequent reels.
Meanwhile some image analysis is getting results into the mind of number 6: