search

hizhangp / yolo_tensorflow

Tensorflow implementation of YOLO, including training and test phase.
MIT License
799 stars 442 forks source link

Why the groundtruth label of 2 bounding box in each cell are the same?? Help plz #77

Open weiaicunzai opened 6 years ago

weiaicunzai commented 6 years ago

Here is the code in loss layer:

boxes = tf.reshape(labels[..., 1:5], [self.batch_size, self.cell_size, self.cell_size, 1, 4])
boxes = tf.tile( boxes, [1, 1, 1, self.boxes_per_cell, 1]) / self.image_size

Can someone please tell me why we have same label for both boxes in the same cell??

If the bouding box center of 2 different objects in the image happens to locate at the same cell, shouldn't we have 2 different box labels for them???