Skip to content

Latest commit

 

History

History
95 lines (76 loc) · 2.2 KB

README.md

File metadata and controls

95 lines (76 loc) · 2.2 KB
Raw

event-library

Library for event-based vision

Setup

Clone the repo:

git clone https://gitlab.iit.it/gscarpellini/event-library

Install event_library:

python setup.py install

Contributing

If you want to contribute to event-library, we suggest to create a virtualenv first. You can use pipenv for that.

git clone https://gitlab.iit.it/gscarpellini/event-library
cd event-library
pipenv install
pre-commit install

Basic guidlines:

  1. Start opening an issue on gitlab assessing the feature you're going to implement or the bug you want to fix.
  2. Fork the project and start coding1! 🔥
  3. We use pre-commit to check that code respect PEP8 and good practices
  4. Submit your pull-request to the main repository :)

Building docs

python -m pip install .
python -m pip install -r requirements/docs.txt
sphinx-build -b html docs/source docs/build

Tools

Generator

Example: conversion from avi files to events

python scripts/generate.py frames_dir={INPUT_DIR} output_dir={OUTPUT_DIR}
upsample=true emulate=true representation=voxel

Tree:

+-- inputdir
|	+-- videodir1
|
+-- outputdir
|   +-- videodir1
|     +-- part_0

Example: conversion from png files to events

python scripts/generate.py frames_dir={FRAME_VIDEO_DIR} output_dir={OUTPUT_DIR}
upsample=true extract=false emulate=true representation=voxel

Each video has a imgs directory, where you put the set of frames. Create a fps.txt file in each video directory where you specify the frame-rate of the video as a single integer number (e.g., 30)

Tree:

+-- inputdir
|     +-- videodir1
|         +-- imgs
|         +-- fps.txt
+-- outputdir
|     +-- videodir1
|         +-- part_0

Visualization

You can visualize a npy events files using the visualize script:

python scripts/visualize.py file_path={YOUR_FILE.npy} representation={REPRESENTATION}

Parameters and help

You can obtain a tool help using python {TOOL}.py --help

Representation:

  • voxelgrid
  • constant-count
  • raw

upsample: if true, upsample frames to higher fps using SuperSloMo model

emulate: if true, create output events files as npy using representation stretegy