display boxes in inference result

Which version of rknn-toolkit you used, there is a drawing code in latest (1.3.0) sample code as below:

1. # Draw result
   
2.     for i in range(0, vaildCnt):
   
3.         if candidateBox[0][i] == -1:
   
4.             continue
   
5. 
   
6.         n = candidateBox[0][i]
   
7. 
   
8.         xmin = max(0.0, min(1.0, predictions[0][n][1])) * INPUT_SIZE
   
9.         ymin = max(0.0, min(1.0, predictions[0][n][0])) * INPUT_SIZE
   
10.         xmax = max(0.0, min(1.0, predictions[0][n][3])) * INPUT_SIZE
    
11.         ymax = max(0.0, min(1.0, predictions[0][n][2])) * INPUT_SIZE
    
12. 
    
13.         # print("%d @ (%d, %d) (%d, %d) score=%f" % (topClassScoreIndex, xmin, ymin, xmax, ymax, topClassScore))
    
14.         cv2.rectangle(orig_img, (int(xmin), int(ymin)), (int(xmax), int(ymax)),
    
15.              (random.random()*255, random.random()*255, random.random()*255), 3)
    
16. 
    
17.     cv2.imwrite("out.jpg", orig_img)
    

复制代码

NG1368 发表于 2020-4-20 23:56
I use from1.2.1 version

You have to download the latest rknn-toolkit from http://t.rock-chips.com/forum.ph ... &extra=page%3D1

And you will find the 1.3.0 has more samples there

In the 1.3.0 sample, coco_label_list.txt is the idx2name file.
You can use it to search the class name

NG1368 发表于 2020-4-23 02:39
can you send me a link of example that use from this code for display boxes?

Try it:

https://rockchips-my.sharepoint. ... %2E3%2E0%2Fexamples

https://rockchips-my.sharepoint. ... bWU9WHZsZEpWZm4xMGc

NG1368 发表于 2020-4-24 03:48
it didn't work and ended up in the same error.

  Other people can work properly ...
Try this one:
https://github.com/rockchip-linux/rknn-toolkit