Details about the loss terms (e.g., weights and thresholding value)

[ghost]

Hello! Wonderful work! I'd love to know more details about the loss terms for its training.

1. what the weight values are for each loss term?
2. The paper says the model was trained using keypoint prior loss from [5]. Did you use the same thresholding values from the original paper [5] or did you change those values, e.g., for the keypoint prior loss, [5] uses the Zt = 0.33 for the depth, and 0.1 for keypoint distance thresholding.

Thank you so much in advance!