search

horizon-research / Real-Time-Spatio-Temporal-LiDAR-Point-Cloud-Compression

This is a compression technique that targets for LiDAR-generated point cloud
126 stars 13 forks source link

运行./pcc_test结果出现Loss rate为负值的情况是为什么 #1

Closed cucrui closed 4 years ago

cucrui commented 4 years ago 
./pcc_test --file ../data/0000000000.bin -p 0.18 -y 0.45 -f binary -l 4 -t 0.5 

../data/0000000000.bin
CURRENT pcloud size: 122626
Range Map: 68, 2004
Time taken by projection: 0.0111554 sec.
Loss rate [PSNR]: -0.00307106 Compression rate: 8.89025 bpp.
Single with fitting_cnts: 6758 with unfitting_cnts: 1759
0000000000.bin
b_mat.bin
coefficients.bin
occ_mat.bin
tile_fit_lengths.bin
unfit_nums.bin
0000000000.bin
b_mat.bin
coefficients.bin
occ_mat.bin
tile_fit_lengths.bin
unfit_nums.bin
Single with fitting_cnts: 6758 with unfitting_cnts: 1759
the x-error rate: 0.0572602
the y-error rate: 0.0579662
the z-error rate: 0.0253382
the error rate: 0.102033
the normal error rate: 0.00980171
the distance error rate: 0.0892009
the normal distance error rate: 0.00882406
0.18 : 0.45
the x-error rate: 4.47336e-07
the y-error rate: 5.77314e-07
the z-error rate: 3.51824e-08
the error rate: 8.61986e-07
the normal error rate: 8.35102e-08
the distance error rate: 2.50039e-07
the normal distance error rate: 2.37455e-08
**********************************************************
Compression rate: 8.89025 bpp
Restored compression rate: -nan bpp
Stream compression rate: -nan bpp
2D projection time: 0.0111554 sec
Average fitting time: 0.112999 sec
Average merging time: -nan sec
Loss rate [PSNR]: -0.00307106
Restored Loss rate [PSNR]: -nan
Merge-Matched Percentage: -nan in -nan
Merge-Unmatched Percentage: -nan in -nan
Fitting Ratio: -nan
YuFengUofR commented 4 years ago

Hi there! Sorry, please ignore the PSNR and loss rate, it is deprecated code, it is used for my paper evaluation originally, but I didn't use it in test code. as long as the code below "**" works. It should be good. your printout looks good to me.

cucrui commented 4 years ago

Hi there! Sorry, please ignore the PSNR and loss rate, it is deprecated code, it is used for my paper evaluation originally, but I didn't use it in test code. as long as the code below "**" works. It should be good. your printout looks good to me.

Thanks for your reply.

YuFengUofR commented 4 years ago

ok, I will close this issue.