README.md

Bumblebee

Bumblebee provides high level components to construct training pipelines for videos conveniently.

• Install
• Motivation
• Our Websites
• Examples
  • A pipeline with basic elements
  • Using Manager API
  • Read limited section of video
  • Iterate frames with frame numbers
  • Iterate frames in batches
• Team
• License

Install

pip install eyecu_bumblebee

Motivation

Everything should be made as simple as possible, but no simpler. - Albert Einstein

Our Websites

EyeCU Vision
EyeCU Future

Examples

A pipeline with basic elements

from bumblebee import *


if __name__ == "__main__":
    
    video_path = "/path/to/video.mp4"

    # Create a source
    file_stream = sources.FileStream(video_path)

    # Add an effect
    goto = effects.GoTo(file_stream)

    END_OF_VIDEO = file_stream.get_duration()
    goto(END_OF_VIDEO)

    # Create a dataset
    single_frame = datasets.Single(file_stream)

    last_frame = single_frame.read()

Using Manager API

from bumblebee import *


if __name__ == "__main__":
    
    # Create a training manager
    manager = managers.BinaryClassification(
        ["path/to/video_dir","path/to/another_dir"],
        ["path/to/labels"]
    )

    number_of_epochs = 300
    
    for epoch,(frame_no,frame,prob) in manager(number_of_epochs):
        # Use data stuff
        ...    

Read limited section of video

from bumblebee import *


if __name__ == "__main__":
  
    video_path = "/path/to/video.mp4"
    start_frame = 35
    end_frame = 40
    
    file_stream = sources.FileStream(video_path)
    
    limited_stream = effects.Start(file_stream,start_frame)
    limited_stream = effects.End(limited_stream,end_frame)

    single_frame = datasets.Single(file_stream)

    for frame in single_frame:
        ...  

Iterate frames with frame numbers

from bumblebee import *


if __name__ == "__main__":
  
    video_path = "/path/to/video.mp4"
    
    file_stream = sources.FileStream(video_path)
    
    single_frame = datasets.Single(file_stream)
    current_frame = effects.CurrentFrame(file_stream)
    
    
    for frame_ind,frame in zip(current_frame,single_frame):
        ...  

Iterate frames in batches

from bumblebee import *


if __name__ == "__main__":
  
    video_path = "/path/to/video.mp4"
    batch_size = 64
    
    file_stream = sources.FileStream(video_path)
    
    batch = datasets.Batch(file_stream, batch_size=batch_size)
    
    for frames in batch:
        ...  

Team

This project is currently developed and maintained by ovuruska.

License

Bumblebee has MIT license. You can find further details in LICENSE.