Unintuitive output of predict_jsons()

[woctezuma]

I was surprised by the output of predict_jsons() when the input image did not contain any face.

The expected behaviour would have been an empty list, so that I could find the number of detected faces as len(annotations).

Instead, there is a list with 1 element, with empty bounding-boxes and a score of -1, due to the following line:

retinaface/retinaface/predict_single.py

Lines 92 to 93 in 29611c7

	if boxes.shape[0] == 0:
	return [{"bbox": [], "score": -1, "landmarks": []}]

I think it is unintuitive and can lead to some confusion for the user.
A more consistent and intuitive behaviour would return annotations as initialized here:

retinaface/retinaface/predict_single.py

Line 66 in 29611c7

annotations: List[Dict[str, Union[List, float]]] = []

Indeed:

• if new faces are detected, they are appended to annotations,
• otherwise, annotations would be returned exactly as it was initialized.

retinaface/retinaface/predict_single.py

Lines 124 to 130 in 29611c7

	annotations += [
	{
	"bbox": bbox.tolist(),
	"score": scores[box_id],
	"landmarks": landmarks[box_id].reshape(-1, 2).tolist(),
	}
	]