Recherche avancée

Sur d’autres sites (16856)

• FFMPEG output to the Exact Folder using Python

  6 août 2021, par Ande Caleb

  i'm working on a simple script using ffmpeg, to reduce the size of a video and add watermark to the video, then move the final output into the compressed folder... this is my script.

  


  the compression works, the watermark works, but the issue i'm having is that the final output is placed in the root folder, and not in the compressed folder... below i my folder structure and my scripts

  


  Folder Structure

  


     rootfolder
    |
    |--media
       |--vids
          |--(video files, mov, mp4s)..
       |--compressed
    |--encode.py


  


  Script (encode.py) file

  


  import os    
import subprocess
from pathlib import Path


dir_path = os.path.dirname(os.path.realpath(__file__))    
vidfile = dir_path + '/media/vids/mv1.mov'    
watermark = dir_path + '/media/watermark.png'
compressed = str(Path.cwd() / '/media/compressed/')

# 1. compress the video and store it in the media out folder

media_out = str(dir_path + "/compressed_mv1s.mov").replace(" ", "\\ ") 
subprocess.run("ffmpeg -i " + vidfile.replace(" ", "\\ ") +
               " -vcodec libx264 -crf 22 " + media_out, shell=True)  

#2.add watermark to the video and move it to the compressed folder 

media_watermarked = str(compressed + '/w_mv1.mov').replace(" ", "\\ ")
subprocess.run("ffmpeg -i " + media_out + " -i " + watermark +
               " -filter_complex \"overlay=main_w-(overlay_w+10) : main_h-(10+overlay_h)\" " + media_watermarked, shell=True)


  


  in summary, compressing the video works, adding watermark works, but the last line, the error is from the media_watermarked variable, i'm not sure what i'm doing wrong but it isn't resolving the folder correctly moving the final output to the folder.. this is the error i get

  


  

  


  Also, how can i run two ffmpeg commands concurrently to compress the video and add watermark at once without doing it seperately.
Thanks.

  


• Trying to convert frames to a video using ffmpeg in a batch script [closed]

  21 octobre 2020, par Katzenwerfer

  I'm doing a script to convert frames into a video.

  


  @echo off
set /p ftv="Want to covert the frames to video (yes or no): "
if %ftv%==yes (echo Ok, initializing frame convertion... & timeout 1 >nul & set /p framerate="Please specify the framerate: " & set /p crf="Please specify a CRF value: " & timeout 1 >nul & ffmpeg -framerate %framerate% -i "%cd%\Interpolated_frames\%%6d.png" -c:v libx264 -preset veryslow -crf %crf% "%cd%\FinalVideo.mp4" & echo Video ready, check the folder for the final video & timeout 1 >nul)


  


  It is suppposed to let you choose the framerate and the crf value, but when running it, ffmpeg gives me Output file #0 does not contain any stream

  


  Dividing it into singular parts like this make it work, but I need to use it with the if

  


  @echo off    
echo Ok, initializing frame convertion...
timeout 1 >nul
set /p framerate="Please specify the framerate: "
set /p crf="Please specify a CRF value: "
timeout 1 >nul & ffmpeg -framerate %framerate% -i "%cd%\Interpolated_frames\%%6d.png" -c:v libx264 -preset veryslow -crf %crf% "%cd%\FinalVideo.mp4"
echo Video ready, check the folder for the final video
timeout 1 >nul


  


• ffmpeg : Converting animated GIF files to video while upscaling produces a file with inaccurate colors

  14 juin 2020, par Metamoran

  I apologize if this is a dumb question, but even after using the search function I have not seen anyone asking about this.

  



  I'm trying to both 
(1) : Convert some low-resolution animated GIF files (pixel art in particular) to video, and
(2) : Upscale them at the same time, using nearest-neighbor to preserve the hard edges.

  



  ffmpeg does everything with no warnings or errors whatsoever, but the end result's colors look off. If I convert without upscaling, the color accuracy is preserved. I have tried both using and NOT using "palettegen", but it does not make a difference. For brevity, I'm only pasting the lines with palettegen in them. The end results are the same either way.

  



  This is what I've been using for upscaling :

  



  ffmpeg -i input.gif -c:v libx264 -b:v 10000K -y -vf "split[s0][s1];[s0]palettegen[p];[s1][p]paletteuse,scale=2*iw:2*ih:flags=neighbor" output.mp4


  



  This is what I used for testing conversion (with no upscaling) :

  



  ffmpeg -i input.gif -c:v libx264 -b:v 10000K -y -vf "split[s0][s1];[s0]palettegen[p];[s1][p]paletteuse" output.mp4


  



  Here's the input file :

  



  Original animated GIF file

  



  Here's a screenshot of how the final video looks if I don't upscale (colors look exactly like in the input file) :

  



  Screenshot - Final - No Upscaling

  



  And here's the result if I upscale (2x Nearest-Neighbor upscaling, with the screenshot downsized 50% to make comparison easier. Colors were not altered during the process.) :

  



  Screenshot - Final - 2x Upscaling (Nearest-Neighbor) [Downsized]

  



  Folder with all relevant files :
Google Drive

  



  Is there something I'm missing ? Or is there some sort of technical limitation, a step that will alter the colors of the video no matter what I try ? I'm not technically inclined, I'd like to know if that is the case. Thank you for your time.