Recherche avancée

Sur d’autres sites (3111)

• What is the Best Approach for Storing and Displaying Video Files as Base64 in HTML for High Performance and Efficiency ? [closed]

  25 août 2023, par Barthez

  I'm currently working on a project where I need to allow users to upload videos, which will then be converted to Base64 and embedded within an HTML file. I'm reaching out to gain a better understanding of the best practices for accomplishing this task while adhering to Stack Overflow guidelines.Here's my plan so far:Video Upload : Users will be able to upload videos through a web interface.Conversion to Base64 : The uploaded video will be converted to Base64 using [specific library/tool].Embedding in HTML : The Base64 encoded video will be embedded within an HTML file using the tag.Before I proceed, I have a few questions:Are there any particular libraries or tools you recommend for converting videos to Base64 efficiently ?What are the potential performance implications of embedding Base64 videos in HTML files, especially considering large video files ?Are there any security concerns I should be aware of when implementing this process ?How can I ensure cross-browser compatibility when embedding these Base64 videos ?Are there any alternatives to this approach that might be more efficient or manageable ?I want to make sure I'm following best practices and avoiding any pitfalls, so any insights, tips would be greatly appreciated. Thank you for your time and assistance !

  


  I attempted to convert an uploaded video to Base64 and embed it within an HTML file.

  


• Extract crisp frames of a video using command line

  11 novembre 2024, par minseong

  How can I use ffmpeg to extract the frames of a video that have no motion blur ?

  


  Using

  


  ffprobe -f lavfi -i "movie=mymovie.mov,blurdetect[out0]" -show_entries tags=lavfi.blur -of default=nw=1 -v quiet


  


  I have learned that the lowest lavfi.blur values of my video are in the range 7-8. For my video, there are 344 lines in this command's output, and 110 of them have values in the range 7-8.

  


  I tried to extract only those frames to images using this command :

  


  ffmpeg -i mymovie.mov -vf "blurdetect,metadata=select:key=lavfi.blur:value=8:function=less"  noblurimages/out%d.png


  


  But many of the output images are repeated, identical images.

  


  There are 168 output images (I was expecting 110), and 124 are identical

  


  md5sum noblurimages/* | sort | awk {'print $2, $1'}  | uniq -f 1 -D | wc -l


  


  Shows that 124 of my output images are identical images.

  


  Out of, I'm assuming, 110 potential motion blur-free frames, I only succeeded to extract 44 unique frames. Something is very wrong with my approach.

  


  How can I get all the motion blur-free frames from a video, and output them as images ?

  


• FFMPEG : invalid data found when processing input, but same input works in FFPLAY

  20 janvier 2023, par SoerenS

  I have a TCP stream with raw 8-bit image data (1280x720) that I receive and display with FFPLAY without problems, but it´s not possible to process the same input with FFMPEG.

  


  This works :

  


  ffplay -i tcp://127.0.0.1:8485\?listen -f rawvideo -pixel_format gray -video_size 1280x720

  


  But when I try to process the same stream with FFMPEG, it always immediatly returns after the first frame with :

  


  tcp://127.0.0.1:8485\?listen: Invalid data found when processing input

  


  I tried the following commands :

  


  ffmpeg -i tcp://127.0.0.1:8485\?listen -f rawvideo -pixel_format gray -video_size 1280x720
ffmpeg -i tcp://127.0.0.1:8485\?listen -f rawvideo -pixel_format gray -video_size 1280x720 -c:v mjpeg -f mjpeg -qscale 6 test.avi

  


  Same behavior under Windows and Ubuntu. FFMPEG and FFPLAY are from the same set of prebuilt binaries. Version command returns N-107067-g5d5a014199-20220609. Unfortunately I have no access to the code of the stream origin. Any idea about a potential root-cause and what I could try ?