Skip to content
All work
Computer Vision

Open-vocabulary detection turns a photo of produce into a bill.

Detection notebook with recorded output. The Flask front end depends on a hosted model endpoint that is no longer available.

YOLO-World LRoboflow inferencesupervisionOpenCVFlask
Detected tomatoes and potatoes with bounding boxes

Overview

Zero-shot detection with YOLO-World: set the class list to the menu items, detect them in a photo, and sum their prices into a bill.

Recorded results

Recorded run
4 classes, bill 46.0
Tomato, Potato, Cucumber, Green capsicum
Source: AI_project (1).ipynb, cells 10 and 11