Question on Time-Varying Scale Factor for Gaussian Motion

[xiao10ma]

Thank you for your great work!
I have a small question regarding the statement in your paper:

To address this issue, we introduce a time dilation factor $\lambda$ to scale the temporal input for each Gaussian point, which is formulated as: $t_s=\lambda_s \cdot t+\lambda_b$

However, during motion, the speed of a Gaussian may vary—it could move faster or slower. Would it be reasonable to keep the coefficients constant in such cases?