niloofarsabouri

Yeah, there are more way to approach it. If you just use OAK as a camera, and do the processing on the Jetson, you could follow and implement something like this square detector.

On the other hand, neural network should also be able to detect this, given the shape and color is very distinct. In that case you would need a relatively good and diverse dataset. I would say you'd need at least 100 images for the first iteration, then train the model. After training, you should inspect the test results you get from the training on "test" data (data not used for training). If performance is not good enough, you should collect and annotate more images and repeat.

      Matija

      Hi and thank you for your response.

      Matija, I have just added my custom model in Oak GUI. By running the custom model I got wrong or false detection.

      The Oak has showed some reactions. but they are not true…decetions are not blue window as I have mentioned before.

      I am exceeding my dataset to 250 images and I have a question, Do you think something else has gone wrong?

      How can I make Oak detect Truly ??

      or this false detection is due to my dataset?

      except increasing the dataset, what are other factors can cause this false detection?

      the image below is the detections by custom model.

      and another question, I have noticed in your custom model folder a file written in python and named by handler… what is handler and how it works and should have to write for my custom models?

      Can you explain it more …

      Thanks in advance.

        Handler is how the bboxes are postprocessed and visualized. CC @jakaskerl to explain in more details.

        For training, how are you training your models? The easiest way to get it working would be to follow one of our tutorials (like YoloV6) here: https://colab.research.google.com/github/luxonis/depthai-ml-training/blob/master/colab-notebooks/YoloV6_training.ipynb. Then you can take the .pt weights and upload them to tools.luxonis.com.

        In Inference section, you can see the inference of the trained model. I'd recommend you check there first. If model is working correctly, then try it on OAK (use tools.luxonis.com to make an export). If the predictions are not OK, try to add more images and quality data and retrained the model.

        I'll let @jakaskerl answer how to add a Yolo to the OAK GUI.

          Matija

          Hi,

          My friend insists on that Oak only woks with YOLO TINY 3, and his reason is in GUI we have detection by yolo tinyv3.

          I have trained my model by Yolo v7. in my model I got nothing and false detection.

          I am now exceeding my data set to 210 images.

          and you have said Inference section…Can you explain it more?

          Thanks alot.

              niloofarsabouri
              Can we first confirm the model behaves as expected, before adding handlers to GUI app.
              Could you try and modify the script for RGB yolo preview. I assume you will be using YoloDetectionNetwork node (since you are using yolov7), so won't need additional on host decoding.
              Anchors and anchor masks are not needed for yolov7 (I think) so you can remove that part. Play around with IOU threshold and confidence.

              Thanks,
              Jaka

                  niloofarsabouri
                  RGB yolo example I sent above will help you find out if the model works properly (pinpointing the issue to either the model or the way the depthai_demo is handling the results).

                  Use the example as is, but change the blob path to your own model blob file. Change the preview size to match the YOLO input size. Remove

                  detectionNetwork.setCoordinateSize(4)
detectionNetwork.setAnchors([10, 14, 23, 27, 37, 58, 81, 82, 135, 169, 344, 319])
detectionNetwork.setAnchorMasks({"side26": [1, 2, 3], "side13": [3, 4, 5]})

                  change other parameters to conform to your NN.

                  Thanks,
                  Jaka

                      jakaskerl

                      Hi again,

                      I divided my dataset into 3 categories: train, test and val.

                      As you see in pictures I pointed the folders, but I got sth wrong… I can not understand…

                      Thanks.

                        jakaskerl

                        Hi

                        what is wrong with this file?

                        I exactly place train, test and valid folder in the training notebook.

                        I am so confused…..

                        Can you explain it more and in detail?

                        I think the my training is not operates.

                          jakaskerl changed the title to Training custom yolo model .

                          Hi @niloofarsabouri
                          Can't really tell what the issue is here. There looks to be a mistake in folder structure and model config. Could you maybe share the .ipynb with outputs.

                          Thanks,
                          Jaka

                          3 months later

                          The Official Guide has some mistake with the folder structure, I came up with the same issue.
                          I modified the folder structure as following, just like other suggestion from different reource.

                          dataset/Image/train

                          dataset/Image/val

                          dataset/Image/test

                          dataset/label/train

                          dataset/label/val

                          dataset/label/test