In the caption of Figure 5, you said "We then compute the IoU with ground truth for each prediction and show a cumulative average of IoU." How is the correspondence between the predicted bounding box and ground truth determined?
Suppose that we got 50 box predictions from the input image and 10 GT boxes, how do I determine the ground truth box for each box?
In the caption of Figure 5, you said "We then compute the IoU with ground truth for each prediction and show a cumulative average of IoU." How is the correspondence between the predicted bounding box and ground truth determined?
Suppose that we got 50 box predictions from the input image and 10 GT boxes, how do I determine the ground truth box for each box?