Recherche avancée

Sur d’autres sites (7508)

• ffmpeg filter complex error ( burning subtitles used overlay filter)

  8 septembre 2020, par jgkim0518

  I try to burn dvb subtitles, based image, on video used ffmpeg overlay filter. but I failed because wrong using filter complex.

  


  It's my command line.

  


  ./ffmpeg -y -hwaccel cuda -hwaccel_output_format cuda -hwaccel_device 0 \
-i input.ts \
-filter_complex "[v:0][s:3]overlay[overlay];[overlay]hwupload_cuda[base];[base]scale_npp=1920:1080[v1];[base]scale_npp=1920:1080[v2];[base]scale_npp=1280:720[v3];[base]scale_npp=720:480[v4];[base]scale_npp=480:360[v5]" \
-map "[v1]" -map 0:a -c:v hevc_nvenc -b:v 6000000 -maxrate 7000000 -bufsize 12000000 -g 15 -c:a libfdk_aac -ar 48000 -ac 2 -pkt_size 128000 -f mpegts test_1.ts \
-map "[v2]" -map 0:a -c:v h264_nvenc -an -b:v 4000000 -maxrate 5000000 -bufsize 8000000 -g 15 -f mpegts test_2.ts \
-map "[v3]" -map 0:a -c:v h264_nvenc -an -b:v 2500000 -maxrate 3500000 -bufsize 5000000 -g 15 -f mpegts test_3.ts \
-map "[v4]" -map 0:a -c:v h264_nvenc -an -b:v 1500000 -maxrate 2500000 -bufsize 3000000 -g 15 -f mpegts test_4.ts \
-map "[v5]" -map 0:a -c:v h264_nvenc -an -b:v 800000 -maxrate 1800000 -bufsize 2000000 -g 15 -f mpegts test_5.ts


  


  but I failed. It is error messages.

  


  Input #0, mpegts, from 'input.ts':
Duration: N/A, start: 22881.964411, bitrate: N/A
  Program 1 
    Metadata:
      service_name    : Service01
      service_provider: FFmpeg
    Stream #0:0[0x100](eng): Audio: ac3 ([129][0][0][0] / 0x0081), 48000 Hz, 5.1(side), fltp, 384 kb/s
    Stream #0:1[0x101](ind): Audio: ac3 ([129][0][0][0] / 0x0081), 48000 Hz, stereo, fltp, 192 kb/s
    Stream #0:2[0x102](zho): Audio: ac3 ([129][0][0][0] / 0x0081), 48000 Hz, stereo, fltp, 192 kb/s
    Stream #0:3[0x103](kho): Audio: ac3 ([129][0][0][0] / 0x0081), 48000 Hz, stereo, fltp, 192 kb/s
    Stream #0:4[0x104]: Video: h264 (High), 1 reference frame ([27][0][0][0] / 0x001B), yuv420p(top first, left), 1920x1080 (1920x1088) [SAR 1:1 DAR 16:9], 25 fps, 50 tbr, 90k tbn, 50 tbc
    Stream #0:5[0x105](CHI): Subtitle: dvb_subtitle ([6][0][0][0] / 0x0006)
    Stream #0:6[0x106](CHS): Subtitle: dvb_subtitle ([6][0][0][0] / 0x0006)
    Stream #0:7[0x107](IND): Subtitle: dvb_subtitle ([6][0][0][0] / 0x0006)
    Stream #0:8[0x108](THA): Subtitle: dvb_subtitle ([6][0][0][0] / 0x0006)
    Stream #0:9[0x109](MAN): Subtitle: dvb_subtitle ([6][0][0][0] / 0x0006)
    Stream #0:10[0x10a](MON): Subtitle: dvb_subtitle ([6][0][0][0] / 0x0006)
    Stream #0:11[0x10b](BUR): Subtitle: dvb_subtitle ([6][0][0][0] / 0x0006)
    Stream #0:12[0x10c](ENG): Subtitle: dvb_subtitle ([6][0][0][0] / 0x0006)
    [mpegts @ 0x47cbd00] Invalid stream specifier: base.
    Last message repeated 17 times
    Stream specifier 'base' in filtergraph description [v:0][s:3]overlay[overlay];[overlay]hwupload_cuda[base];[base]scale_npp=1920:1080[v1];[base]scale_npp=1920:1080[v2];[base]scale_npp=1280:720[v3];[base]scale_npp=720:480[v4];[base]scale_npp=480:360[v5] matches no streams.


  


  My plan is this.


  


  How to do burn subtitle on video using filter complex, ffmpeg from this structure ?

  


• Is it possible to send a temporary slate (image or video) into a running Azure Live Event RTMP-stream ?

  15 novembre 2020, par Brian Frisch

  I'm currently building a video streaming app which leverages Azure Media Services Live Events.

  


  It consists of :

  


  

  1. a mobile app that can stream live video and.

  


  2. a web client that plays the live event video.

  


  3. a producer screen with controls to start and stop the web client access to the video.

  


  4. a server that handles various operations around the entire system

  


  


  It's working very well, but I would like to add a feature that would enable the producer to add some elegance to the experience. Therefore I'm trying to get my head around how I can enable the producer be able to switch the incoming source of the stream to a pre-recorded video or event a still image at any point during the recording, and also to switch back to live-video. A kill-switch of some kind, that would cover waiting-time if there's technical difficulties on the set, and it could also be used for pre-/post-roll branding slates when introing and outroing a video event. I would like this source switch to be embedded in the video stream (also so it would be possible to get this into the final video-product if I need it in an archive for later playback)

  


  I'm trying to do it in a way where the producer can set a timestamp for when the video override should come in, and when it should stop. The I want to have my server respond to these timestamps and send the instructions over RTMP to the Azure Live Event. Is it possible to send such an instruction ("Hey, play this video-bit/show this image in the stream for x-seconds") in the RTMP-protocol ? I've tried to figure it out, and I've read about SCTE-35 markers and such, but I have not been able to find any examples on how to do it, so I'm a bit stuck.

  


  My plan-B is to make it possible to stream an image from the mobile application that already handles the live video-stream, but I'm initially targeting an architecture where the mobile app is unaware of anything else than live streaming, and this override switch should preferably be handled by the server, which is a firebase functions setup.

  


  If you are able to see other ways of doing it, I'm all ears.

  


  I've already tried to build a ffmpeg method that listens to updates to the producer-set state, and then streams an image to the same RTMP-url that the video goes to from the mobile app. But it only works when the live video isn't already streaming - it seems like I cannot take over a RTMP-stream when it's already running.

  


• How to create readable file stream from multiple files

  9 septembre 2020, par lowcrawler

  Related to question : Add specific image files to ffmpeg video with fluent-ffmpeg

  


  Have array of file names that reside on network shares. Would like to turn them into a timelapse video with a process on a nodejs server. I plan to use ffmpeg to do this. FFMPEG doesn't allow file lists as input (as far as I can tell). They do, however, allow a 'readable file stream'.

  


  Is there a mechanism to create a readable file stream from an array of filenames ?

  


  The command they use in the fluent-ffmpeg example is : var command = ffmpeg(fs.createReadStream('/path/to/file.avi'));

  


  ... but obviously, this is only a single file.