Katsuya Hyodo

@acad1234 First, please publish the IR conversion script and .pb you used. I do not know what is the problem even if only an error message is presented. The IR...

@acad1234 Please try the following. However, I am still under operation verification. **https://github.com/PINTO0309/OpenVINO-YoloV3/wiki/Reference-repository#conversion-success** **keras2tensorflow** **https://github.com/amir-abdi/keras_to_tensorflow/tree/24c4f8d715148a42690e8cd089bfe5eeeed3a8ea**

Thank you very much! I used an aspect ratio of 1:1 until two weeks ago, but I found that the accuracy dropped significantly. This is not a big problem with...

Thank you again. The logic you presented is exactly the one I used up to two weeks ago. btw, "cv2.dnn.blobFromImage" is very simple and easy to handle, but it is...

As you say, there was a bug in the calculation of the drawing position and drawing size of the bounding box.

@superJerryLee It is not a problem of the program. The essence is that the accuracy of the tiny model is very bad. Editing "yolo_v3_tiny_changed.json" does not change anything. Change "Threshold...

@superJerryLee I will tell you the same thing again. Adjust the "threshold" described in the program. For example: ```python objects = ParseYOLOV3Output(output, m_input_size, m_input_size, camera_height, camera_width, 0.2, objects) ``` ```...

@superJerryLee Ummm... Same issue #2 , #6 tiny-YoloV3 is very poor accuracy... **https://pjreddie.com/darknet/yolo/** Is it possible to share your model file with me? I would like to investigate on Saturday...

@superJerryLee I do not know if the position where the frame is displayed is correct. The number of frames displayed is not infinite. - **Program** openvino_tiny-yolov3_MultiStick_test.py - **Points changed according...

@superJerryLee >1. Are you converting from yoloV3-tiny.weights to FP16? Yes. I used the model file that you provided to me.　(frozen_tiny_yolo_v3.bin / frozen_tiny_yolo_v3.xml) >2. If you adjust Threshould =0.2, will...