วีดีโอ

Wow ARKit works much better than I thought!

오 김교수님 재밌는거 하고 지내시는군요! :)

Which SLAM is this ? Is it using an iPhone or any other 3d camera ?

Where does the errors come from? Do you think it is one big localization errors or mild errors gradulately accumulated?

Thank you for the video. I use the 3DScanner app for interior lidar scans. I utilize the data from it to create 3D architectural models in Blender. Do you have any suggestions to streamline this process? My biggest issue is that the walls are not straight. Is there a way to straighten the walls, ceiling, and floor using CloudCompare or any other software? Thank you!

으으 아이폰으로 기변해야되나 생각들게 하는 영상이군요! 임용 축하드립니다 교수님 :)

Great

I understand absolutely nothing but it looks good so good job

감사합니다!

Thank you so much, but how can I convert the map to pcd?

goofiest vid I've seen for a while

Do you have sample data and code for converting a polar image to a Cartesian image in the video? Thank you very much!

와우.. 예상은 했지만 역시 거울 뒤로 다른 세계를 만들어버리는군요.. mirror detection을 할 수 있는 네트워크로 recon한다면 어떨지 궁금합니다 ㅎㅎ

what exactly is a false loop ?

Any idea on how to run it with live camera feed?

can i have your code? it helps me a lot :))

Hello dear Kim. I am working on a Drone Project and am trying to implement Fast-LIO. Currently all get while playing my bag is the LIDAR pointcloud under "Cloud registered" but no actual map is created. I have a topic named "Lasermap" but its doesnt publish any data. I looked at on of your old videos where you implemented Fast-LIO and alot of the topics you have arent present for me. If it wouldnt be to much of a hassle for you I would greatly appreciate your help. Best regard - Marcel

Amazing work. What would you suggest to go with in a new autonomous indoor navigation project? I'm looking into SC-LIO-SAM, but many other of your repositories seem relevant too. I currently work in Humble, with a ros1_bridge for SLAM - as a temporary solution. LIO-SAM has it's ROS2 version - did you consider updating SC-LIO-SAM for ROS2? I wonder if we should stick with SC-LIO-SAM, or consider going with another algorithm (with 6DoF IMU if possible :)). Thank you!

와.. 핸드폰에서 리얼타임으로 바로 리컨이 되는건가요?

DOES THIS WORK WITH LIVOX MID 360?

Hello, your work is so impressive. I'm a student and have been learning about your work recently. I'm curious to know if LT-SLAM can be integrated with SC-LIO, and if it's challenging to do so.

Hello, I have been learning scan context recently, and I would like to know how you drew this depth map. I hope you can reply, which will be greatly appreciated

Boa tarde! Não consigo ativar o painel de propriedades! Poderia me ajudar? Só está visível o console e árvore de entidades. Me instruíram para que eu olhasse na parte inferior esquerda para alternar entre botões mas, também não está visível na parte inferior. Já desinstalei e reinstalei o cloudcompare e não tive sucesso...

I am using Livox_ros_driver2, and hence getting error, please help me

will this work with Livox Mid 360 lidar?

Hello, did you measure the CPU usage of this integrated method in comparison to the bare Fast-LIO2? I wonder if it will be usable on TX2 or similar devices. Thank you for your efforts and insight!

Haha, coincidence I've just installed LM Studio a couple of hours ago and started playing with it. Tried Wizardcoder 34B but i didn't manage to make it work, I'll try the 13B tomorrow.

Will this work on a velodyne lidar?

슬램kr 오프라인 발표 듣고 바로 구독합니다ㅎㅎ

hello I have onequestion, so if i have in my robot, odometry and sequence scans, i can implement FAST-LIO by modyifing the topics which is subscribed to, and it will perform the slam with my data?? or do you have any advice? thanks

좋은 영상 감사드립니다!

영상 감사드립니다.!

if we use this LiDAR then the Moving object segmentation algorithm would be useless ?

Are you using ROS Noetic or Humble ?

감사합니다.

How you are getting the velocity?

Can i get the code and dataset name ?

lol, why even bother make such a video?

Impressive point cloud edit skill. Can you share some good tutorials for learning them?

Impressive!

nice job!

멋지내요 굿~~~ sketchfab 이 사이트는 편집, 업로드 등이 무료 인가요?

nice try. if the scene can be converted into mesh or surfel then the visualization may be better than point clouds 😀

Would be nice if I could take the photos into meshroom with good quality. Quick Scans would be nice. Tricky part is scanning a room, can't make large changes or take to many photos.

Hi, this is amazing work!! Kudos! Would you mind sharing your github repo?

Hi! I have try your method. However, the map in the rviz is not updated although the closed loop condition have been met. Is there are any parameter that I need to add to enable the updated map?

Is there an open-source ROS driver for the A65 that you used?

no sound!

오 재밌는 실험이네요! "이 환경은 비관성계이다" 라는 것을 염두해서 vehicle state도 같이 추정하여 보정하는 방법도 좋겠네요

Hi @gskim Nice demonstration!! do you know how to create .g2o files for custom dataset? I would like to test on a custom dataset Thanks in Advance!