Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/eyecuvision/bumblebee
Video Processing API
https://github.com/eyecuvision/bumblebee
computer-vision data-processing numpy opencv torch video-processing-pipeline
Last synced: 4 months ago
JSON representation
Video Processing API
- Host: GitHub
- URL: https://github.com/eyecuvision/bumblebee
- Owner: eyecuvision
- License: mit
- Created: 2020-12-17T16:38:41.000Z (over 3 years ago)
- Default Branch: main
- Last Pushed: 2021-04-24T12:53:57.000Z (about 3 years ago)
- Last Synced: 2024-03-01T05:36:12.580Z (4 months ago)
- Topics: computer-vision, data-processing, numpy, opencv, torch, video-processing-pipeline
- Language: Python
- Homepage:
- Size: 45.9 MB
- Stars: 8
- Watchers: 1
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Lists
- awesome-from-stars - eyecuvision/bumblebee
README
# Bumblebee
[](https://pypi.python.org/pypi/eyecu_bumblebee)
[](https://pepy.tech/project/eyecu-bumblebee) \
Bumblebee provides high level components to construct training pipelines for videos conveniently.
- [Install](#install)
- [Motivation](#motto)
- [Our Websites](#our-websites)
- [Examples](#examples)
- [A pipeline with basic elements](#a-pipeline-with-basic-elements)
- [Using Manager API](#using-manager-api)
- [Read limited section of video](#read--limited-section-of-video)
- [Iterate frames with frame numbers](#iterate-frames-with-frame-numbers)
- [Iterate frames in batches](#iterate-frames-in-batches)
- [Team](#team)
- [License](#license)
## Install
```
pip install eyecu_bumblebee
```
## Motivation
Everything should be made as simple as possible, but no simpler. - Albert Einstein
## Our Websites
[EyeCU Vision](https://eyecuvision.com/) \
[EyeCU Future](https://eyecufuture.com/)
## Examples
### A pipeline with basic elements
```python
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
```python
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
```python
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
```python
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
```python
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](https://github.com/ovuruska).
## License
Bumblebee has MIT license. You can find further details in [LICENSE](LICENSE).