Skip to content

Latest commit

 

History

History
56 lines (43 loc) · 2.02 KB

README.md

File metadata and controls

56 lines (43 loc) · 2.02 KB
Raw

VOCALExplore

This repository contains the code for VOCALExplore. The code for the experiments in the technical report is on the experiments branch. The main branch extends the experimental implementation to serve as the backend for an exploration interface.

Demonstration

Clicking on the preview image will navigate to a demonstration video on the UWDB website.

Project structure

vfe/ contains the implementation of VOCALExplore. vfe/api houses the various managers and task scheduler, and the rest of the subdirectories implement utilities.

Docker

Build the dockerfile:

docker build -t vocalexplore/vocalexplore -f docker/Dockerfile .

Copy .env and modify any ports or paths. Pass this env file to start_docker below. Make sure that the thumbnail directory matches the directory that will be specified in the server's configuration yaml file.

  • SERVER_PORT: The port that the server will be available on.
  • THUMBNAIL_DIR: The directory where thumbnails will stored.
  • THUMBNAIL_PORT: The port where thumbnails can be accessed.
  • VIDEO_DIR: The directory where videos are stored.
  • VIDEO_PORT: The port where videos can be accessed.

When running the frontend, copy the contents of .env into config.py.

Start the docker container in the background:

./start_docker.sh

Args:
  -e <path to env file>, or default .env
  -n <docker container name>, or default vocalexplore

Server

The server can be started by running

python server/server.py

Arguments:
  -c/--config-file <path to config file>

Sample config files can be found in server/configs. The db_dir is the directory where VOCALExplore's storage manager will store its data. If thumbnail_dir is specified, VOCALExplore will extract thumbnail images from loaded videos and store them in this directory.