thm

video post-processing for The History Makers

makevideos

this script takes the raw video captures delivered by THM personnel and:

1. concatenates the < 4GB files into 1 long file

2. transcodes that file to flv, mp4, and mpeg

3. embeds timecode and watermarks where appropriate

4. hashmoves them to their destiantions

5. triggers script to embed those hashes into a Filemaker db named PBCore_Catalog

makevideos also checks to make sure that everything is plugged in and that all necessary files (like watermarks) are in their expected locations.

makevideos also uses MediaConch to check the format compliance of every input and output video

makevideos can also be run manually by cd'ing into the repo directory (look for that in the config.txt file) and running "python makevideos.py"

##ffmpeg strings these aren't implemented quite as they are written here, everything in brackets is a variable for example, but if you wanted to make each of these derivatives with ffmpeg, this is what you would use:

concatenate

ffmpeg -f concat -i concat.txt -map 0:v -map 0:a -c:v copy -c:a copy -timecode 00:00:00:00 rawconcat.mov

route channel 2 audio out and re-wrap

ffmpeg -i rawconcat.mov -map 0:a:1 -map -0:d -map -0:v -c:a copy rawconcat-as2.mov

ffmpeg -i rawconcat.mov -i rawconcat-as2.mov -map 0:v -map 0:a:0 -map 1:a:0 -map 0:d -c copy concat.mov

flv

ffmpeg -i concat.mov -i "/Volumes/G-SPEED Q/Titan-HD/HM/Watermark/Video WaterMark_White.png" -filter_complex "scale=320:180,overlay=0:0;[0:a:0][0:a:1]amerge=inputs=2[a]" -c:v libx264 -preset fast -b:v 700k -r 29.97 -pix_fmt yuv420p -c:a aac -ac 2 -map 0:v -map "[a]" -timecode 00:00:00:00 -threads 0 [accessionNumber.flv]

mpeg

ffmpeg -i concat.mov -target ntsc-dvd -filter_complex "[0:a:0][0:a:1]amerge=inputs=2[a]" -b:v 5000k -vtag xvid -vf "drawtext=fontfile='/Library/Fonts/Arial.ttf': timecode='00\:00\:00\:00': r=29.97: x=(w-tw)/2: y=h-(2*lh): fontcolor=white: fontsize=72: box=1: boxcolor=0x00000099,scale=720:480" -map 0:v -map "[a]" -ac 2 -threads 0 [accessionNumber.mpeg]

mp4

ffmpeg -i concat.mov -c:v mpeg4 -b:v 372k -pix_fmt yuv420p -r 29.97 -vf "drawtext=fontfile='/Library/Fonts/Arial.ttf': timecode='00\:00\:00\:00': r=29.97: x=(w-tw)/2: y=h-(2*lh): fontcolor=white: fontsize=72: box=1: boxcolor=0x00000099,scale=420:270" -filter_complex "[0:a:0][0:a:1]amerge=inputs=2[a]" -c:a aac -ar 44100 -ac 2 -map 0:v -map "[a]" -threads 0 [accessionNumber.mp4]

*test input

if you want to generate a test input file for this situation here's how.

first, make a video file in the usual way

ffmpeg -f lavfi -i "testsrc=duration=10:size=1920x1080:rate=29.97" -c:v mpeg2video -timecode 00:00:00.0 [vout].mov

then make an audio file and wrap it in a mov

ffmpeg -f lavfi -i "sine=frequency=1000:sample_rate=48000:duration=10" -c:a pcm_s24be [aout].mov

then wrap the video file with the audio file mapped to two different streams, with timecode track

ffmpeg -i [vout].mov -i [aout].mov -c:v copy -c:a pcm_s24be -map 0:v:0 -map 0:a:0 -map 1:a:0 -timecode 00:00:00:00 [out].mov