Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Next revision
Previous revision
public:gsoc:2018:achie27 [2018/08/10 06:59]
achie27 wrote introduction
public:gsoc:2018:achie27 [2018/08/17 18:17] (current)
achie27 formatting
Line 1: Line 1:
 ===== Introduction ===== ===== Introduction =====
  
-Videos contain plethora of contextual information. For example, in a movie there are fighting scenes, sentimental scenes, romantic scenes, and many others. In a cricket match, there are wickets, sixes, fours et cetera. With the advent of the data-driven age, amateurs, researchers,​ and organisations alike require some specific part of this contextual information for their needs; maybe for creating a highlights reel of a sports match or mining data from movies for their machine learning models. This makes parts of certain types of videos very useful. FabBits tries to automate ​that.+Videos contain plethora of contextual information. For example, in a movie there are fighting scenes, sentimental scenes, romantic scenes, and many others. In a cricket match, there are wickets, sixes, fours et cetera. With the advent of the data-driven age, amateurs, researchers,​ and organisations alike require some specific part of this contextual information for their needs; maybe for creating a highlights reel of a sports match or mining data from movies for their machine learning models. This makes parts of certain types of videos very useful. FabBits tries to automate ​finding them\\  
 +Following are the things it will be able to detect - 
 + 
 +  * Action sequences in movies/​shows - ✅\\ 
 +  * Summary of movies/​shows - ✅\\ 
 +  * Actor-specific scenes in movies/​shows - ✅\\ 
 +  * Jokes in sitcoms - ✅\\ 
 +  * Slo-mos in Sports - ❌\\ 
 +  * Goals in Soccer - ✅\\ 
 +  * Goal misses in Soccer - ⭕\\ 
 +  * Three pointers in Basketball - ✅\\
  
 ===== Links ===== ===== Links =====
 Project repo - [[https://​github.com/​achie27/​FabBits|github.com/​achie27/​FabBits]] \\  Project repo - [[https://​github.com/​achie27/​FabBits|github.com/​achie27/​FabBits]] \\ 
-Blog posts - [[http://​www.medium.com/​@achie27 | medium.com/​@achie27]]+Blog posts - [[http://​www.medium.com/​@achie27 | medium.com/​@achie27]] ​\\ 
 +Samples - [[ https://​drive.google.com/​drive/​folders/​1lsrLW0c7Pq4FYpWZjrCqKs3RFTwL-1ub | Drive folder]] 
 +===== Requirements ===== 
 +You need the following things to run FabBits-\\  
 +1. [[ https://​www.python.org/​download/​releases/​3.0/​ | Python3]] \\  
 +2. [[ https://​opencv.org | OpenCV ]]- Used for image and video processing \\  
 +3. [[ https://​zulko.github.io/​moviepy/​ | Moviepy ]] - Used for video editing and audio processing \\  
 +4. [[ https://​www.riverbankcomputing.com/​software/​pyqt/​intro | PyQt5 ]] - Used to make the GUI \\  
 +5. [[ https://​scipy.org | Scipy ]] - Used for audio processing \\  
 +6. [[ https://​github.com/​sirfz/​tesserocr | Tesserocr]] - Used for, well, OCR \\ 
 +7. [[ https://​pillow.readthedocs.io/​en/​latest/​ | Pillow ]] - Used to preprocess images for OCR \\ 
 +The python dependencies can be installed by running - \\  
 + 
 +  pip3 install scipy 
 +  pip3 install opencv-python  
 +  pip3 install moviepy 
 +  pip3 install pyqt5 
 +  pip3 install Pillow 
 +  pip3 install tesserocr 
 + 
 +or if you are the Anaconda kind - 
 + 
 +  conda install -c conda-forge scipy 
 +  conda install -c conda-forge opencv  
 +  conda install -c conda-forge moviepy  
 +  conda install -c anaconda pyqt 
 +  conda install -c conda-forge pillow 
 +  conda install -c simonflueckiger tesserocr 
 +   
 +===== Usage ===== 
 +Run the main GUI by - 
 +  python3 main.py 
 + 
 +To find your FabBit of choice - 
 +  * Click ''​MOVIES''​ or ''​SPORTS''​ button for their respective use-cases 
 +  * Select the use-case from the sidebar 
 +    * A pop-up dialog will ask for the actor if actor-specific scene was chosen 
 +  * Click on ''​Choose File''​ to select the input video  
 +  * Click on ''​Find FabBits''​ 
 +  * Move the slider in the blue areas, which are the extracted FabBits, and play the video 
 +  * Click on ''​Save FabBits''​ to save the extracted stuff into a video file 
 + 
 +You can also run the respective files of use-cases to get their FabBit, like - 
 +<​code>​python3 goal_detector.py soccer_match.mp4</​code>​
  
 +===== References =====
 +All the references can be found listed in the repository'​s [[https://​github.com/​achie27/​FabBits#​readme|readme]].
  
  • public/gsoc/2018/achie27.1533884352.txt.gz
  • Last modified: 2018/08/10 06:59
  • by achie27