0

I have a old VCD, and I want to enhance it using AI tech. my questions are:

  1. how to know the origin frame rate of this VCD(dat file)?
  2. how to split this VCD into images with the raw frame rate? ( I don't want to miss one frame )
  3. does google has the tech to enhance the old images?
  4. once all the images are enhanced , how to re-combine these into a new video?

If these are some AI tech to repair/convert the entire VCD file, it's also a good choice.

thanks a lot!

Siwei
  • 19,858
  • 7
  • 75
  • 95

1 Answers1

0

OK let me answer my question and you can vote for close as your wish, bosses.

  1. how to know the origin frame rate of this VCD(dat file)?

use this command:

ffmpeg -i target.dat -ss 00:01:00 -t 10 images/out-%08d.png

this will split the 1st minutes of your video, and you will see its origin frame rate

Input #0, mpeg, from 'MUSIC01-马礼堂养气功VCD.DAT':
  Duration: 00:58:04.32, start: 0.653333, bitrate: 1411 kb/s
    Stream #0:0[0x1e0]: Video: mpeg1video, yuv420p(tv), 352x288 [SAR 178:163 DAR 1958:1467], 1120 kb/s, 25 fps, 25 tbr, 90k tbn, 25 tbc
    Stream #0:1[0x1c0]: Audio: mp2, 44100 Hz, stereo, s16p, 224 kb/s

it's 25 fps, 352x288 each frame image size.

all the old VCD has the same fps ( 25 )

does google has the tech to enhance the old images?
once all the images are enhanced , how to re-combine these into a new video?
  1. how to split this VCD into images with the raw frame rate?
ffmpeg -i target.DAT images/out-%08d.png

this will produce images like:

out-00000001.png
out-00000002.png
out-00000003.png
...
  1. does google has the tech to enhance the old images?

I don't know. will open a new thread/question.

  1. once all the images are enhanced , how to re-combine these into a new video?
ffmpeg -loglevel trace -f image2 -r 25 -pattern_type sequence -i 'tmp_enlarged_image_folder/vcd3-%08d.png' -vcodec libx264 test.mp4

refer to convert from jpg to mp4 by ffmpeg

Siwei
  • 19,858
  • 7
  • 75
  • 95