Skip to content

Latest commit

 

History

History
62 lines (45 loc) · 1.47 KB

README.MD

File metadata and controls

62 lines (45 loc) · 1.47 KB
Raw

ClearView

Just another ML for detecting and classifing objects on images packed in simple web gallery. This project is a part of the python course at the University DSW.

Description

The project is a image gallery that allows the user to upload images and then the system will detect and classify the objects on the image. The project is using the EfficientDet model for object detection and classification, Flask for the web framework, and Bulma for the front-end framework, font-awesome for icons.

TODO

  • Choose a model [Detectron2 + CLIP]
  • Choose a database [Postgres]
  • Choose web framework [Flask]
  • Choose a front-end framework [Bootstrap]
  • Dockerize the project
  • Create report for the project (working)
  • Create a menu
  • Create a gallery
  • Create a upload form
  • Create search form
  • Implement database initialization
  • Implement (some of kind?) API
  • Implement images scanning
  • Test models
  • Implement models (soon will be working on it)

Instruction | NOT WORKING YET |

  1. Clone the repository
  2. Run in your terminal
    pip install -r requirements.txt

then

    python website/scripts/create_db.py --load-dummy-data

as a result you should see

Database created successfully.
20 dummy photos added to the database.

and instance/gallery.db file should also be created

  1. Run in your terminal
python website/app.py

and visit with your browser:

http://127.0.0.1:5000