How to reduce number of classes in YOLOv3 files?

by

For easy and simple way using COCO dataset, follow these steps :

  • Modify (or copy for backup) the coco.names file in darknet\data\coco.names
  • Delete all other classes except car
  • Modify your cfg file (e.g. yolov3.cfg), change the 3 classes on line 610, 696, 783 from 80 to 1
  • Change the 3 filters in cfg file on line 603, 689, 776 from 255 to 18 (derived from (classes+5)x3)
  • Run the detector ./darknet detector test cfg/coco.data cfg/yolov3.cfg yolov3.weights data/your_image.jpg

For more advance way using COCO dataset you can use this repo to create yolo datasets based on voc, coco or open images. https://github.com/holger-prause/yolo_utils .
Also refer to this : How can I download a specific part of Coco Dataset?

Would be great if you can train YOLO model using your own dataset. There are so many tutorial on the internet of how to build your own dataset. Like this, this, this or this.

Note : reducing number of classes won’t make your inference speed faster. By reducing classes, you will detect less object and somehow will probably make your program run faster if you do post-processing for each detection.

More Related Contents:

  1. opencv c++ Edge detection to find edge that the top of it is white and the bottom is black
  2. How to verify the correctness of calibration of a webcam?
  3. opencv drawing a 2d histogram
  4. Computing camera pose with homography matrix based on 4 coplanar points
  5. How to capture the desktop in OpenCV (ie. turn a bitmap into a Mat)?
  6. OpenCV C++/Obj-C: Advanced square detection
  7. Displaying stitched images together without cutoff using warpAffine
  8. Configuring an c++ OpenCV project with Cmake
  9. Could not find module FindOpenCV.cmake ( Error in configuration process)
  10. 2 usb cameras not working with opencv
  11. sift = cv2.xfeatures2d.SIFT_create() not working even though have contrib installed
  12. Square detection doesn’t find squares
  13. How to crop away convexity defects?
  14. CMake Error: Variables are set to NOTFOUND
  15. MinAreaRect angles – Unsure about the angle returned
  16. Why cv2.so missing after opencv installed?
  17. How to load png images with 4 channels?
  18. OpenCV as a static library (cmake options)
  19. Filling gaps in shape edges
  20. Can’t access webcam with OpenCV
  21. cv2.imshow() function is opening a window that always says not responding – python opencv
  22. How to detect region of large # of white pixels using OpenCV?
  23. How to use Bazel to build project uses OpenCV [closed]
  24. Are there any fast alternatives to SURF and SIFT for scale-invariant feature extraction?
  25. OpenCV with GigE Vision Cameras [closed]
  26. How is the camera coordinate system in OpenCV oriented?
  27. How to detect ellipses in image without using fitEllipse() in opencv?
  28. How can I get ellipse coefficient from fitEllipse function of OpenCV?
  29. OpenCV window always on top
  30. Package opencv was not found in the pkg-config search path

Leave a Comment