I am collecting images for object detection from a simulator. Is there any software to labeling bounding boxes (bx, by,bh,bw) for each image and generate it in TensorFlow format to use it as (train_x,train_y)? A GUI app would be much better as it would make it easier to make an accurate bounding box.
Your help is appreciated. Thanks
A well known tool for labeling bounding box is labelImg: https://github.com/tzutalin/labelImg
To convert your dataset in TensorFlow format (I guess you mean tfrecord format), you can use a python script as this one: https://github.com/tensorflow/models/blob/master/research/object_detection/dataset_tools/create_pet_tf_record.py