Object Video Annotation

OVA is a simple GUI for making video annotations of object recognition experiments

Some code are adapted from https://github.com/Wahl-lab/EXPLORE

Instructions

Basically, you upload your behavior video, select the position of the object, give them a name and a key label ('p' reserved).

After that, you will see a frame just to remember the key label of each object in case you have many objects.

After pressing finish, you will see the video and pressing each key label will select the frames that the animal explores the object.

Installation

• clone repo
• create conda env

  conda env create -f environment.yml

• activate env

  conda activate OVA

• run script

  python main.py