Recherche avancée

Sur d’autres sites (7380)

• ffmpeg : Crop webm file with circular .png mask [closed]

  26 mars 2021, par Beneos Battlemaps

  I tried everything for hours and can get it working by myself. I want to create animated Pen&Paper tokens for virtuale tabletops. I have a .webm video file with 720x720 pixel showing an animation of a render file created out of a png sequence via

  


  ffmpeg -framerate 24 -f image2 -i face.%04d.png -c:v libvpx-vp9 -pix_fmt yuva420p face1.mp4

  


  See here : https://webmshare.com/play/ZxZa0

  


  Now i want to overlay this .webm with an circle image as frame :

  


  overlay

  


  and crop the black part so its transparent in the end result as well as keeping the transparent area in the center for the video itself.

  


  So the endresult is from this :

  


  Face1

  


  to this :

  


  Face2

  


  Can you help me out with that ? If its easier i can split the steps, making first the alpha mask crop via png and adding the circle in it later.

  


  Beste regards in advance
-Ben

  


• How to set RTSP mode to TCP using Xuggler

  26 octobre 2013, par Anurag Joshi

  I am working with a Sanyo VCC HD2300P IP camera and trying to capture the frame from the live stream on the camera. I have another camera that provides an rtsp stream and the Xuggler code works fantastically on that. However, on the sanyo camera I get the error below

  could not open stream: rtsp://admin:admin@192.168.0.3:554/VideoInput/1/h264/1

  I use this stream on iSpy and VLC and am able to get the stream. Using the functionality from these 2 programs I am also able to capture a frame from the stream.
  The URL for the other camera that works well is rtsp ://admin:123456@192.168.0.246/mpeg4cif. So on the face of it I don't see a reason why it should not work.

  Any help will be highly appreciated.

  EDIT : I searched and experimented a lot. Earlier I was using the inbuilt support for Sanyo cameras in iSpy using which the camera stream was visible. However, when I tried to directly provide the URL for the H.264 stream, the stream would not work unless I changed RTSP mode to TCP. This gave me a hint that probably I need to call the IContainer.open method in a manner that uses RTSP over TCP.
  Googling about it threw some possible solutions like

  using rtsp://192.168.0.3:554/VideoInput/1/h264/1?tcp

  It says that this would force the stream to open on TCP. However, nothing has changed for me. I still get the same error.

  EDIT2 : So essentially it boils down to how do I set the RTSP transport mode to TCP ? By default the transport mode for RTSP is UDP.

• How to avoid critical security issues introduced by ffmpeg ? which seems to be a dependency of DeepFace for running opencv in linux environment

  29 août 2022, par Shubham Patel

  I was trying to run DeepFace in docker and when I ran the container I got an error related to OpenCV.

  


  Following online suggestions, I downloaded the FFmpeg package and it resolved the OpenCV error and everything was working fine inside the container.

  


  I ran a docker scan for checking the security issues and it highlighted 4 critical severity issues introduced through FFmpeg package

  


  Can anyone help me how to avoid these security issues ?

  


  Below is the content of the docker file :

  


  
RUN apt-get update
RUN apt-get install ffmpeg  -y

RUN pip install flask flask_cors deepface numpy pillow flask_wtf

WORKDIR /app
COPY . /app

EXPOSE 84
CMD ["python","app.py"]


  


  Below is the result of the docker scan which uses snyk, NOTE : I am just providing High and Critical Severity issues.

  


    Description: Out-of-bounds Write
  Info: https://snyk.io/vuln/SNYK-DEBIAN11-GDKPIXBUF-2960116
  Introduced through: ffmpeg@7:4.3.4-0+deb11u1, gdk-pixbuf/libgdk-pixbuf2.0-bin@2.42.2+dfsg-1, librsvg/librsvg2-common@2.50.3+dfsg-1
  From: ffmpeg@7:4.3.4-0+deb11u1 > ffmpeg/libavcodec58@7:4.3.4-0+deb11u1 > librsvg/librsvg2-2@2.50.3+dfsg-1 > gdk-pixbuf/libgdk-pixbuf-2.0-0@2.42.2+dfsg-1 > gdk-pixbuf/libgdk-pixbuf2.0-common@2.42.2+dfsg-1
  From: gdk-pixbuf/libgdk-pixbuf2.0-bin@2.42.2+dfsg-1 > gdk-pixbuf/libgdk-pixbuf-2.0-0@2.42.2+dfsg-1
  From: librsvg/librsvg2-common@2.50.3+dfsg-1 > gdk-pixbuf/libgdk-pixbuf-2.0-0@2.42.2+dfsg-1
  and 2 more...
  Image layer: 'apt-get install ffmpeg -y'

✗ High severity vulnerability found in aom/libaom0
  Description: Out-of-bounds Write
  Info: https://snyk.io/vuln/SNYK-DEBIAN11-AOM-1085722
  Introduced through: ffmpeg@7:4.3.4-0+deb11u1
  From: ffmpeg@7:4.3.4-0+deb11u1 > ffmpeg/libavcodec58@7:4.3.4-0+deb11u1 > aom/libaom0@1.0.0.errata1-3
  Image layer: 'apt-get install ffmpeg -y'

✗ Critical severity vulnerability found in zlib/zlib1g
  Description: Out-of-bounds Write
  Info: https://snyk.io/vuln/SNYK-DEBIAN11-ZLIB-2976151
  Introduced through: meta-common-packages@meta
  From: meta-common-packages@meta > zlib/zlib1g@1:1.2.11.dfsg-2+deb11u1
  Image layer: Introduced by your base image (python:3.9.13-slim)

✗ Critical severity vulnerability found in aom/libaom0
  Description: Release of Invalid Pointer or Reference
  Info: https://snyk.io/vuln/SNYK-DEBIAN11-AOM-1290331
  Introduced through: ffmpeg@7:4.3.4-0+deb11u1
  From: ffmpeg@7:4.3.4-0+deb11u1 > ffmpeg/libavcodec58@7:4.3.4-0+deb11u1 > aom/libaom0@1.0.0.errata1-3
  Image layer: 'apt-get install ffmpeg -y'

✗ Critical severity vulnerability found in aom/libaom0
  Description: Use After Free
  Info: https://snyk.io/vuln/SNYK-DEBIAN11-AOM-1298721
  Introduced through: ffmpeg@7:4.3.4-0+deb11u1
  From: ffmpeg@7:4.3.4-0+deb11u1 > ffmpeg/libavcodec58@7:4.3.4-0+deb11u1 > aom/libaom0@1.0.0.errata1-3
  Image layer: 'apt-get install ffmpeg -y'

✗ Critical severity vulnerability found in aom/libaom0
  Description: Buffer Overflow
  Info: https://snyk.io/vuln/SNYK-DEBIAN11-AOM-1300249
  Introduced through: ffmpeg@7:4.3.4-0+deb11u1
  From: ffmpeg@7:4.3.4-0+deb11u1 > ffmpeg/libavcodec58@7:4.3.4-0+deb11u1 > aom/libaom0@1.0.0.errata1-3
  Image layer: 'apt-get install ffmpeg -y'



Organization:      16082204
Package manager:   deb
Target file:       Dockerfile
Project name:      docker-image|face-verification-v2
Docker image:      face-verification-v2
Platform:          linux/amd64
Base image:        python:3.9.13-slim
Licenses:          enabled

Tested 314 dependencies for known issues, found 120 issues.

According to our scan, you are currently using the most secure version of the selected base image```