alliedel/videofeatures
Folders and files
| Name | Name | Last commit date | ||
|---|---|---|---|---|
Repository files navigation
% First author and code base owner: Allie Del Giorno % This code implements the discriminative framework for the A. Del Giorno, J.A. Bagnell, M. Hebert ECCV 2016 paper "A Discriminative Framework for Anomaly Detection in Large Videos" % If you have issues running the code or plan to use it for research purposes, please contact me: adelgior@cs.cmu.edu . % ===== Installation ===== % ffmpeg must be installed on your system (to unpack the videos). % ===== Usage ===== % == Demo % Run the script 'demo.m': >> matlab -r demo This script should run and terminate with a print statement "Demo success!" and a time. On my (laptop) machine, it takes about 15 minutes. The greatest proportion of time is spent 'unpacking' the videos. % == Modifications % You'll want to modify the script code/scripts/runscript.m as a starting point. % Parameter modifications can be made in Configure.m (make your own configuration) % The set of all parameters and their defaults are listed in code/src/configure/DefaultPars.m . You can change most of these parameters in Configure.m . % ===== Functionality ===== % This code takes a video file (.mp4, .avi) and returns a .train file (libsvm format) with features from the video. % It will NOT generate anomaly scores for videos. To generate your own features (or put them into the correct format for this code to use), please download the latest version of: ___________________ . % One set of features is provided for comparison in: data/input/features/Avenue/03. % ===== System Requirements ===== % Linux with MATLAB % This code has been tested on Ubuntu 12.04. The 'core' of the implementation is in C++, though some functionality/wrapping is provided in MATLAB. % ===== Other notes ===== % This code was originally combined with the anomaly detection code before I decided to split it for usability, so if you see remnants of that code, my apologies!