README.md 1.29 KB
Newer Older
Matteo Spanio's avatar
Matteo Spanio committed
1
2
3
4
# Workflow

## Description

Matteo's avatar
Matteo committed
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
```mermaid
sequenceDiagram
 participant main as Main Thread
 participant audio as Audio Analyser
 participant video as Video Analyser
 participant tapeRestoration as Tape Audio Restoration
 participant tapeClassifier as Tape Irregularity Classifier
 participant packager as Packager
 participant data as Working Directory

 main ->> audio: GET /irregularityFile1.json
 audio ->> data: Save irregularityFile1
 main ->> video: GET /irregularityFile1.json
 main ->> video: GET /irregularityFile2.json
 video ->> data: Save IrregularityFiles
 main ->> audio: GET /irregularityFile2.json
 Note right of data: All IrregularityFiles,<br/>images and audio<br/>have been stored

 packager ->> data: Store everything

```

Matteo's avatar
update    
Matteo committed
27
## Getting started
Matteo Spanio's avatar
Matteo Spanio committed
28
29
30
31
32
33

## Installation

## Usage

## Support
Matteo's avatar
update    
Matteo committed
34
If you require additional information or have any problem running the software, please contact us at:
Matteo's avatar
Matteo committed
35
* Matteo Spanio (dev2@audioinnova.com)
Matteo Spanio's avatar
Matteo Spanio committed
36
37

## Authors and acknowledgment
Matteo's avatar
update    
Matteo committed
38
39
40
41
42
This project was developed by:
* Matteo Spanio (Audio Innova);
* Nadir Dalla Pozza (University of Padova);
* Niccolò Pretto (University of Padova);
* Sergio Canazza (University of Padova).
Matteo Spanio's avatar
Matteo Spanio committed
43

Matteo's avatar
update    
Matteo committed
44
This project takes advantage of docker and docker-compose.
Matteo Spanio's avatar
Matteo Spanio committed
45

Matteo's avatar
update    
Matteo committed
46
47
## License
This project is licensed with [GNU GPL v3.0](https://www.gnu.org/licenses/gpl-3.0.html).