If I have a Poisson point process $\mathcal{X}$ of density $\lambda$ on the Euclidean plane $\mathbb{R}^2$, with the Euclidean metric taking pairs of points to the Euclidean distance,
$$ \operatorname{dist} (\langle x_1, y_1 \rangle, \langle x_2, y_2 \rangle) = \sqrt{(x_1 -x_2)^2-(y_1-y_2)^2} $$
can I allow the metric to depend on time, in such a way that it smoothly (or otherwise) approaches the hyperbolic metric as time traverses the closed unit interval $[0,1]$?
So, initially, $\mathcal{X}$ sees Euclidean space, and finally hyperbolic space,
$$ \operatorname{dist} (\langle x_1, y_1 \rangle, \langle x_2, y_2 \rangle) = \operatorname{arcosh} \left( \cosh y_1 \cosh (x_2 - x_1) \cosh y_2 - \sinh y_1 \sinh y_2 \right) $$ with a number (perhaps a finite set) of "negatively curved" spaces in between. Only a small region would need this property, not the whole of $\mathbb{R}^2$.
The reason for this is it would affect the structure of a random graph or simplical complex built on the points (e.g. the degree distribution). Is this possible?
Note: I understand Ricci flow is similar to this, but where the metric satisfies a partial differential equation.