I have trained yolov3 on a custom dataset following the implementation from (https://github.com/AlexeyAB/darknet) repo. I'm getting bounding boxes for this test image but not getting labels for that. Can you please help me fix this issue?
Related
I have a set of icons and a screen recording, the icons are not annotated and have no bounding boxes, they are just png icons with image-level labels eg: "instagram", "facebook", "chrome".
The task is to find the icons within the screen recording and draw a bounding box around them, given the above prerequistes.
My idea of approach so far is:
Use selective search to find ROIs
Use a CNN to classify the regions
Filter out non-icons regions
Draw bounding boxes around positive labelled ROIs
Use resulting screen images with bounding boxes to train a FAST R-CNN
but I am stuck at step 2, I have no idea on how to train the CNN with the image-level labelled icons.
If I make a dataset of all the possible icon-images, with no background or context informations, is it possible to train the CNN to answer the question "Does the ROI includes a possible icon?"
I'm wondering about image resizing and then the intuitive bounding box resizing that would follow that.
For instance, when I use a 640x640 image in my dataset, and the model has a fixed_shape_resizer of 320x320, will the original bounding box be scaled down to match the smaller 320x320 size?
Yes, Tensorflow will automatically resize the bounding box to match the smaller input size.
Here is a Link to the code that changes the bounding box sizes.
I'm working on an AndroidApp with TensorFlow Lite to process semantic segmentation.
https://www.tensorflow.org/lite/api_docs/java/org/tensorflow/lite/InterpreterApi
Also I'm using U2net model and processing images which are already masked like below.
enter image description here
Problem is masked area could be circle or elliptical shape, so it can't be done with normal process.
Throwing this image into interpreter results in getting just same image in which circle area is white color.
So I have to cut out the circle and throw it, but only the circle area is included.
Throwing cut-out image does't work, because it's out of shape.
The arrays must be square just like normal images.
I counld't find any lead to solve this problem.
Any advice would be appriciated.
Thank you.
I'm training an object detection model with Tensorflow (fine tuning from ssd_mobilenet_v2_320x320_coco17_tpu-8) and monitor the training task with tensorboard. I was expecting in the Images tab of tensorboard that images that are displayed would show a bounding box. What I see though is only images with an orange line drawn above the picture (the same orange that I expect for the bounding box). Am I missing something? Am I right when I say that a bounding box should appear or not? Picture of what I see is joined. Any help greatly appreciated.
I have trained YOLOv3 darknet model on my custom dataset. First I have cloned the https://github.com/kriyeng/darknet/ repository in google colab and then run the whole code in google colab. But I want to change the color of the bounding boxes in the output video. Please suggest how to do it. Thanks in advance.
Here is a solution in C++ but you can apply it to python as well. Just change the values in Scalar. In the example below it is set to the color red.
//Draw a rectangle displaying the bounding box
rectangle(frame, Point(left, top), Point(right, bottom), Scalar(0, 0, 255));