https://github.com/barleo01/yoloobjectdetector
Yolo Object detector with Opencv and Qt
https://github.com/barleo01/yoloobjectdetector
Last synced: 5 months ago
JSON representation
Yolo Object detector with Opencv and Qt
- Host: GitHub
- URL: https://github.com/barleo01/yoloobjectdetector
- Owner: barleo01
- Created: 2019-05-29T07:49:04.000Z (almost 6 years ago)
- Default Branch: master
- Last Pushed: 2019-05-29T11:50:20.000Z (almost 6 years ago)
- Last Synced: 2024-08-02T01:25:18.892Z (9 months ago)
- Language: C++
- Homepage:
- Size: 1.16 MB
- Stars: 6
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
- awesome-yolo-object-detection - barleo01/yoloobjectdetector
- awesome-yolo-object-detection - barleo01/yoloobjectdetector
README
# Object detector with Opencv and Qt
The pupose of this application is to capture video from a camera, apply a YOLO Object detector and display it on a simple Qt Gui.
## Processing
The processing of the Yolo Detector (10 CNN) took ~0.5 seconds on 7 cores. In order to have a fluid application, not all images could be processed. I used thread 1 and a timer to grab the images every milliseconds.
On another thread, the processing were done. Once finished it sends the output of the layer to the thread 1 and received a new image. I used QThread for the implementation
## Data needed
To work, the application need the trained weights of the neural network. You can find there:
*https://pjreddie.com/media/files/yolov3.weights
The configuration of the neural network can be found here:
*https://github.com/pjreddie/darknet/blob/master/cfg/yolov3.cfg
*https://github.com/pjreddie/darknet/blob/master/data/coco.names
All three files must be placed in a folder called "data" in the cmake directory
