Hi, @yifita. I find your great work recently.
But I am a little confused about the gradient of the regularization terms. In the paper, 4.2 Alternating normal and point update, you said the point and normal are updated by the gradient of the regularization terms. I find the code in https://github.com/yifita/DSS/blob/d96260c8c0b926ba2fd43d82eb3e0afd970a046a/DSS/training/losses.py#L182
maybe related.
The Let pi denote a point in question and pk denote one point in its neighborhood. So the Lr and Lk are constant with know pi, pk.
How to calculate the gradient with respect to pk?
Thanks,
Hi, @yifita. I find your great work recently. But I am a little confused about the gradient of the regularization terms. In the paper, 4.2 Alternating normal and point update, you said the point and normal are updated by the gradient of the regularization terms. I find the code in https://github.com/yifita/DSS/blob/d96260c8c0b926ba2fd43d82eb3e0afd970a046a/DSS/training/losses.py#L182 maybe related. The Let pi denote a point in question and pk denote one point in its neighborhood. So the Lr and Lk are constant with know pi, pk. How to calculate the gradient with respect to pk? Thanks,