https://github.com/mbari-org/sightwire
https://github.com/mbari-org/sightwire
Last synced: 7 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/mbari-org/sightwire
- Owner: mbari-org
- License: apache-2.0
- Created: 2024-05-29T16:34:54.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2024-05-30T19:16:53.000Z (over 1 year ago)
- Last Synced: 2025-06-10T09:58:30.780Z (7 months ago)
- Language: Python
- Size: 2.53 MB
- Stars: 0
- Watchers: 5
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
## Sightwire
sightwire is a tool for setting up a database and managing workflows for media collected for the MBARI CoMPAS project.
The general flow is to create a project, upload media, and then run models for detection, classification, etc. on the media.
This is a work-in-progress and more operations will be added as needed. Contact D.Cline dcline@mbari.org for more information.
A version of this is currently live at http://mantis.shore.mbari.org/9/project-detail? **ONLY WORKS in a CHROME BROWSER**
Username: **guest**
Password: **mbariguest**
## Overview
## Requirements
For development, you will need:
* [Python >=3.9](https://www.python.org/downloads/)
* [Anaconda](https://www.anaconda.com/) or [Miniconda](https://docs.conda.io/en/latest/miniconda.html) is recommended, but not required
* [ffmpeg](https://ffmpeg.org/)
For deployment, you will need:
* [Docker](https://docs.docker.com/get-docker/)
* [Docker Compose](https://docs.docker.com/compose/install/)
* [ffmpeg](https://ffmpeg.org/)
## Development setup
## 1. Install dependencies and activate the environment
```bash
conda env create
conda activate sightwire
```
The command line interface can be run with:
```bash
python sightwire
```
which should print something like:
```bash
Usage: sightwire [OPTIONS] COMMAND [ARGS]...
Run CoMPAS data workflows
Options:
-V, --version Show the version and exit.
-h, --help Show this message and exit.
Commands:
convert Commands related to converting data
create-video Create mp4 video files from images
database Commands related to database management
extract-log Extract timestamp, depth, position, etc.
init Initialize the database
load Commands related to loading data
```
## 2. Mount the compas data directory to e.g. /Volumes/compas or /mnt/compas
Note that you can also use a local directory. Whatever directory you use, that is used in the scripts to launch the database and load data.
Examples:
Linux
```shell
sudo mount -t nfs titan.shore.mbari.org:/data/compas /mnt/compas
```
Mac
```shell
mount smb://titan.shore.mbari.org/compas /mnt/compas
```
## 3. Setup and run the database
See the [database setup](./docs/database_setup.md) instructions for details.
## 4. Load some test data
```bash
./scripts/load_image.sh /Volumes/compas
```
The data can now be viewed in the UI at
http://localhost:8080 and be downloaded programmatically through the API.
For example, to query all media for the platform MINI_ROV and camera FLIR
```python
# Simple example on how to query media with attributes
import os
import tator
# export TATOR_HOST=http://localhost:8080
# export TATOR_TOKEN=15a0f486cde0f09329b09f3274de1b962c2273a4
# export TATOE_PROJECT=902204-CoMPAS
api = tator.get_api(os.environ['TATOR_HOST'], os.environ['TATOR_TOKEN'])
projects = api.get_project_list()
project_name = os.environ['902204-CoMPAS']
project = [p for p in projects if p.name == project_name][0]
# These can also be programmatically queried
mono_image_type = 41 # mono
project_id = 9
# Query for the platform MINI_ROV and camera FLIR
attributes =["$platform::MINI_ROV", "$camera::FLIR"]
# Get all mono images with the above attributes
images = api.get_media_list(project=project_id, type=mono_image_type, attribute=attributes)
print(f'Found {len(images)} images with attributes {attributes}')
```