update parameter sigma about probreg

[cbenitez81]

Sure thing, I haven't had time but I started another filter approach, as soon is tested, I'll let you know

…

On Thu, Oct 31, 2019, 11:22 neka-nat ***@***.***> wrote: I corrected the calculation of simga2. 06013f2 <06013f2> Thanks to this, calculation accuracy has improved. Please let me know if there are any improvements. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#7?email_source=notifications&email_token=AFQEAE2P2T545S2FTUU5KSLQRLZ43A5CNFSM4JCBJ56KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOECYFSZQ#issuecomment-548428134>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AFQEAE24VGBLWQV3IOE3LEDQRLZ43ANCNFSM4JCBJ56A> .

[neka-nat]

Hi @cbenitez81,

Thank you for good suggestion!
I checked the paper and the original code.
You must be right.

Will you send a pull request if possible?

[cbenitez81]

Let me make another couple of changes if I have time next week. I think that we can improve the speed if we run the filter once for the reference cloud and prevent the hash table of growing for the model cloud.

[neka-nat]

Sounds great.
I'm looking forward to the improvement.

[neka-nat]

I corrected the calculation of simga2.
06013f2
Thanks to this, calculation accuracy has improved.
Please let me know if there are any improvements.