ICP & Point Cloud Registration - Part 2: Unknown Data Association (Cyrill Stachniss, 2021)

[@familywu3869]

Excellent lecture! Thank you very much Professor!

[@Shaharsarshalom1]

Thank you so much for this lecture, you are my idol!; This lecture is like superman on execution

[@denispotapov9828]

Thank you for your lectures! It was very helpful.

[@mrc-kit83]

Great talk! Thank U very much!

[@gaurishgangwar]

Thanks for the lecture, Professor. The slides available on your website differ a bit from this lecture, they don't have slides corresponding to 62-65 in the video (where you mention about moving car alongside another car). Will it be possible for you to share the latest slides?

[@gaurishgangwar]

Found it. The order of slides is different but it's there.

[@kapilrawal7870]

This is awesome!, Thanks for the video

[@soroushmaleki6336]

Great video! Also loving the SLAM course. Thanks :)

[@ChrisOffner]

Fantastic lecture again, thank you so much. Are the slides available as PDF somewhere?

[@CyrillStachniss]

Yes, please send me an email I am am happy to share PDFs and PPTX files.

[@alejandrovilla22921]

Hello! Are these different sampling methods available as built-in functions in some Python libraries? Regards.

[@Snouzz-Play]

I NEED HELP ! pls if you kno how to write an code on c# to add some point cloud (ply) Images to one 3d Image so pls help me! Can give you some Steam-Card for you help! i use Intel RealSense d415 Cam, programm wors but it didnt march images well and it cant do real live3d merge...

[@moomoo2214]

I think there's an error in basic ICP algorithm's pseudocode. It should be x_bar_n = R(x_bar_n - x_0) + y_0

[@mariashoukat3923]

Great video sir, can you plz make a video on multiway point cloud registration

[@IngvardVenterai]

using aruco or similar markers to calibrate a multiple camera setup should make ICP simple and fast I believe, would yoy agree ?

[@CyrillStachniss]

Cameras don’t measure the distance to the obstacles, so you cannot apply I CP easily here.

[@lesterknome]

How does ICP with unknown data association compare to optimal transport or Wasserstein distance based alignment methods?

[@CyrillStachniss]

I cannot comment on that because I never tried alignment methods based on optimal transport or Wasserstein distance myself.

[@lesterknome]

@@CyrillStachniss Thanks for your response, I am just learning about ICP :). I am wondering if ICP can handle situations where 1-1 point correspondences do not exist? I.e. different i.i.d. samples from unaligned distributions?

[@bartsimpson208]

@@lesterknome it doesn't require 1-1 correspondence. Also, check out point-to-plane variation

[@nelsonjoppi]

thank you!!

[@frizlang69]

Excellente lecture: could you share the slides?

[@CyrillStachniss]

Sure. Send me an email

[@PhucNguyen-to2mp]

Thanks for Useful lectures, Would it be possible for me to get the pdfs for 3 lectures?

[@CyrillStachniss]

Sure, send me an email and you can get all the PPTX and PDF files

[@敏杨-p1f]

Why does it say that your mailbox does not exist

[@CyrillStachniss]

It does exist…