Open luzzou opened 4 years ago
hi, I have a question about the camera loss.
It seem that the result of tf.trace(m_sq) is not consistent with trace(K\timesK^T) since tf.trace is not a batch operation.
tf.trace(m_sq)
loss_mat = tf.reshape((2 / tf.trace(m_sq)), [-1, 1, 1])*m_sq - tf.eye(2)
Looking forward to your reply.
hi, I have a question about the camera loss.
It seem that the result of
tf.trace(m_sq)is not consistent with trace(K\timesK^T) since tf.trace is not a batch operation.loss_mat = tf.reshape((2 / tf.trace(m_sq)), [-1, 1, 1])*m_sq - tf.eye(2)Looking forward to your reply.