Recherche avancée

Sur d’autres sites (6966)

• Creating RTSP stream in Java - not serving a file

  9 mai 2016, par user2959589

  I’ve done quite a bit of searching and have not found any clear answers. I would like to create a RTSP stream in my Java server app. This stream isn’t just sending out an existing file. No, I want it to stream some images generated from within Java. For example, I might generate a spinning cube, or a graph that updates based on some input (number of queries on my server, data from a sensor, etc), or something like that.

  Essentially I’ll be creating a series of BufferedImages (perhaps 10 per second) and I need to stream them out as RTSP with reasonable video compression. As an advanced aspect, I might want to add some sound after I have the basics working, such as "beep" when the graph goes over some value.

  I’ve seen all the various tool options. There are Java bindings for FFMPEG, VLC, and gstreamer, and perhaps others. But I haven’t seen anyone using them to do what I am proposing here.

  One obvious way to do this is to write the BufferedImages to files, then use ffmpeg to turn those into a video format file of some kind, and then serve that file. That would work, but I need something that’s efficient and low latency (near real-time), not writing out files and converting them.

  Any ideas ? Thank you !

• Python - Extract frames from video without OpenCV

  30 septembre 2019, par bpinaya

  for semantic segmentation inference I’d like to extract frames of a video and run them through a network without using OpenCV’s VideoCapture as something like (answered in this question) :

  vidcap = cv2.VideoCapture('testvideo.mp4')

  the reason being that I don’t want to install that package which is a bit heavy only to get frames from video.

  What I currently do is getting the frames from a video with ffmpeg with :

  ffmpeg -i testvideo.mp4 frames/%05d.png

  And then load the frames, perform inference, and save the results. I’ve seen packages like ffmpeg-python which have examples for how to load the video into numpy arrays, which I haven’t managed to get working.

• To create an android app where I am able to add color filters to a recorded video

  27 novembre 2014, par Ramesh Kannan

  I’m trying to create an app where I am able to add color filters to a recorded video.

  I’ve done research and I can’t piece it all together. I’ve looked into using OpenCV to record a video. The problem is, when I use the sample APK in OpenCV it trigger out the camera and color effects are applied simultaneously then stored as a picture but what I need is after recording the video to apply color effects in the preview mode. I don’t know how to implement OpenCV in my android application.

  I’ve also heard about "ffmpeg" which is an open-source platform for recording, converting, playing and streaming video and audio but I didn’t get any idea to implement it to my project.

  Lastly, I’ve looked into OpenCV. Tried using OpenCV, but I haven’t had much either.

  Do you have steps / procedures / code / example on integrating video filters (FFMPEG) on Android for the following needs.

  • I used to record the video,
  • then dividing it as frames and apply filters.