Closed

RTSP client in C/C++ (GNU) including decoding of H.264 video

This project was awarded to edsiper for $800 AUD.

Get free quotes for a project like this
Employer working
Awarded to:
Project Budget
$250 - $750 AUD
Total Bids
2
Project Description

Hi there, I require a program to acquire the [url removed, login to view] video stream from an IP camera via RTSP

Required is the following:
==
- write a C / C++ (gcc / g++ - Linux RHEL6.4) program using appropriate free libraries
- connect via TCP to the camera, authenticate and request the pre-configured RTSP stream URL (send: DESCRIBE rtsp://$cam_ip:$cam_port/video RTSP/1.0\n" etc. - URL should be configurable
- authenticate via user / password (configurable, in variable)
- needs to limit network bandwidth to the camera to 64, 128 kbps or 256 kbps (configurable, in variable)
- needs to be able to save the [url removed, login to view] video and audio streams (if any) to 5 MB max. chunks (sequential file name or timestamp as name)
- needs to decode the [url removed, login to view] video stream using a suitable free library to jpeg images and save single jpeg images at the frame rate defined (1-16 fps configurable) in the file system (sequential file name or timestamp)
- needs to respond to authentication failures, wrong URL, network timeouts etc. in a safe manner (no segfault or crash)
- needs to run indefinitely (demonstrated for at least 24 hours without crash and memory leaks
- the first image should be available after 4-6 seconds max.
- needs to use the least amount of CPU possible
- can use any free libraries (live555, openRTSP, openav, ffmpeg etc.) as long as they are easily / freely available in RHEL 6.4 64 bit
- remote 24h network access to 2 different demo cameras supplied
- any more questions please let me know prior to posting your bid

Cheers,

Looking to make some money?

  • Set your budget and the timeframe
  • Outline your proposal
  • Get paid for your work

Hire Freelancers who also bid on this project

    • Forbes
    • The New York Times
    • Time
    • Wall Street Journal
    • Times Online