search

BichenWuUCB / SqueezeSeg

Implementation of SqueezeSeg, convolutional neural networks for LiDAR point clout segmentation
BSD 2-Clause "Simplified" License
566 stars 239 forks source link

INPUT_MEAN and INPUT_STD, how the values come? #34

Open chunhuaqiushi1989 opened 5 years ago

chunhuaqiushi1989 commented 5 years ago

hello, all my friends, in the code: norm_lidar = (lidar - self._mc.INPUT_MEAN) / self._mc.INPUT_STD

x, y, z, intensity, distance

mc.INPUT_MEAN = np.array([[[10.88, 0.23, -1.04, 0.21, 12.12]]]) mc.INPUT_STD = np.array([[[11.47, 6.91, 0.86, 0.16, 12.32]]])

And I don't understand how the values(INPUT_MEAN and INPUT_STD) are from?

Meranre commented 5 years ago

i think it's mean of {x,y,z,intensity , distance } and variance of the same data , you can calculate them with your own data if you want , here they have already calculate them i think

Zhen-ao commented 4 years ago

i think it's mean of {x,y,z,intensity , distance } and variance of the same data , you can calculate them with your own data if you want , here they have already calculate them i think

@Meranre Hi~ I also have this problem. This is the value I calculated:

#x, y, z, intensity, distance
mc.INPUT_MEAN = np.array([[[-0.1023, 0.4952, -1.0545, 0.2877, 11.7127]]])
mc.INPUT_STD = np.array([[[12.2958, 9.4287, 0.86, 0.1450, 10.24]]])

I calculate the mean of all x, y, and z in the data set. Especially the value of x=-0.1023 I calculated is quite different from the value providedx= 10.88. How did you solve it?