[FFmpeg-trac] #9184(avfilter:open): pngdec output frame that has double image

[FFmpeg]

#9184: pngdec output frame that has double image
-------------------------------------+-------------------------------------
             Reporter:  akoryagin    |                    Owner:  (none)
                 Type:  defect       |                   Status:  open
             Priority:  important    |                Component:  avfilter
              Version:  unspecified  |               Resolution:
             Keywords:  artefact on  |               Blocked By:
  the video                          |
             Blocking:               |  Reproduced by developer:  0
Analyzed by developer:  0            |
-------------------------------------+-------------------------------------
Comment (by Pljushevij):

 Replying to [comment:22 wenbin,chen]:
 > Hi Pljushevij. Sorry, I forget that there are patches not accepted. You
 can visit the repo: https://github.com/intel-media-ci/cartwheel-ffmpeg and
 apply all the patches in "patches" directory. That is the latest update
 for ffmpeg-qsv and ffmpeg-vaapi.


 Thank you. All works fantastic, but...
 If I am using concat function, when first video finishing and starting
 second video ffmpeg showing error:


 {{{
 Impossible to convert between the formats supported by the filter
 'Parsed_overlay_vaapi_5' and the filter 'auto_scaler_0'
 Error reinitializing filters!
 Failed to inject frame into filter network: Function not implemented
 Error while processing the decoded data for stream #0:0
 }}}

 Both video in HEVC format, but resolution is small different.
 When I was using SW overlay (CPU) this problem not exits. Command (SW
 overlay) is here:


 {{{
 ffmpeg -hwaccel qsv -hwaccel_output_format qsv -qsv_device
 /dev/dri/renderD128 -re -f concat -safe 0 -i "files.txt" -loop 1 -i
 "logo.png" -filter_complex
 "[1]lut=a=val*0.5[a];[0:v]scale=1920:-1:force_original_aspect_ratio=decrease,pad=1920:1080
 :(ow-iw)/2:(oh-ih)/2[bg];[bg][a]overlay=(main_w-overlay_w-1):(main_h-
 overlay_h-1)" -map 0:v:0 -map 0:a:0 -c:v hevc_qsv -preset veryfast -b:v 7M
 -maxrate 7M -c:a eac3 -sn -f hls -hls_list_size 5 -hls_flags
 delete_segments -y stream.m3u8
 }}}

 But this command more load CPU. Your first code with QSV has the same
 problem with concat.
 A was attached ffmpegTest.zip there is file for testing. Two video, logo,
 and files.txt.
 Can you please test it on your side?
-- 
Ticket URL: <https://trac.ffmpeg.org/ticket/9184#comment:23>
FFmpeg <https://ffmpeg.org>
FFmpeg issue tracker