AlexeyAB / darknet

YOLOv4 / Scaled-YOLOv4 / YOLO - Neural Networks for Object Detection (Windows and Linux version of Darknet )
http://pjreddie.com/darknet/
Other
21.71k stars 7.96k forks source link

Output not saving in json format #6208

Open saikrishnadas opened 4 years ago

saikrishnadas commented 4 years ago

Hey, I need to save my image ouput in json format.

Code i ran: ( the same way it was mentioned in the page)

./darknet detect cfg/custom-yolov4-detector.cfg backup/custom-yolov4-detector_final.weights data/test_image.jpg -dont-show -out -out result.json < data/train.txt

But no ouput named result.json is saved.

It works with txt file but not with .json file

@AlexeyAB

stephanecharette commented 4 years ago

Code i ran: ( the same way it was mentioned in the page)

./darknet detect cfg/custom-yolov4-detector.cfg backup/custom-yolov4-detector_final.weights data/test_image.jpg -dont-show -out -out result.json < data/train.txt

Note that what you wrote is not like the example on the page.

saikrishnadas commented 4 years ago

Hey, @stephanecharette Thanks for the response. But how can I save the result to json for a single image instead of a batch of image paths written into data/something.txt

Because using the batch of images method happens to have problem while giving multiple call:

 161 yolo
[yolo] params: iou loss: ciou (4), iou_norm: 0.07, cls_norm: 1.00, scale_x_y: 1.05
nms_kind: greedynms (1), beta = 0.600000
Total BFLOPS 59.701
avg_outputs = 492276
 Allocate additional workspace_size = 52.43 MB
Loading weights from backup/custom-yolov4-detector_final.weights...
 seen 64, trained: 1920 K-images (30 Kilo-batches_64)
Done! Loaded 162 layers from weights-file
Enter Image Path:

It stops there.

So, is there any possible was to save result in json for one image instead of putting path in the .txt file

@AlexeyAB

stephanecharette commented 4 years ago

I use DarkHelp to get my json output. You can find out about it here: https://www.ccoderun.ca/darkhelp/api/ShellScripting.html