This project is using COCO dataset and YOLOV8 for object detection. For distance estimation, it is using the ratios of bounding boxes and coordinates of each. BlindAid is built to replace the traditional white cane used by visually impaired community. The functionality is to detect objects in nearby radius and inform the user using audio feedback. The project uses Yolov8 for object detection and ratios of bounding boxes to estimate the distance of objects. It can detect 80 classes of commonly used objects.
rafaym1/BlindAid
Folders and files
| Name | Name | Last commit date | ||
|---|---|---|---|---|