Why we need to correct scale so that the nearest depth is at a little more than 1.0?

[MMMazart]

I don't understand why we need to correct scale so that the nearest depth is at a little more than 1.0。
` near_original = self.bounds.min()

    scale_factor = near_original*0.75 # 0.75 is the default parameter# the nearest depth is at 1/0.75=1.33
    self.bounds /= scale_factor
    self.poses[..., 3] /= scale_factor`

Could you explain？thank you