unmannedlab / lidarnet Goto Github PK
View Code? Open in Web Editor NEWLiDARNet: A Boundary-Aware Domain Adaptation Model for Point Cloud Semantic Segmentation
Home Page: https://unmannedlab.github.io/research/LiDARNet
LiDARNet: A Boundary-Aware Domain Adaptation Model for Point Cloud Semantic Segmentation
Home Page: https://unmannedlab.github.io/research/LiDARNet
@maskjp @maskjptamu, I have recently read your pre-print for Image/Lidar feature matching (Contrastive Learning of Features between Images and LiDAR). I was wondering if your lab is planning to release the code at some point? I would like to test your method on some airborne data (images/Lidar) that I would like to register.
Hi, Peng Jiang.
Could you please share your code with me? I want to utilize your method on some other datasets. And I guarantee that the code will never be exposed to the public and will not be used for any commercial purposes.
Please help.......
Hey!
Thanks for the great work and releasing the code and dataset. I realized that there are a couple of minor issues in the poses.txt
files. I was wondering if you can help me fix them?
poses.txt
is 1 or 2 lines fewer than the number of velodyne scans in the same sequence. The is the case '00', '02', '04', '05', '06', '07', '08', '09', '10', '11', '13', '14', '16', '17', '18', '19', '20', '22', '23', '24', '25', '26', '27', '28', '29', '30', '31', '33', '34', '35', '36', '37', '38', '39', '40', '41', '42', '43', '44', '45', '46', '47', '48', '49', '51', '53', '55', '57', '59', '61', '63', '65', '67', '68', '69', '70'
poses.txt
has very large values and seems to be wrong. This is the case for '16', '19', '59', '68', '69', '70'
How would you suggest to fix these issues?
Hi, Peng Jiang. Could you please share the code with me? I want to see some specific details of the code and learn from it. Thank you!
Hi, Peng Jiang.
Firstly, thank you for your brilliant work!
But I have a question. In Tab. I, there is an object class. I can't find object class in your paper, so I want to know which classes are turned into object class in your paper.
Please help......
Hi @maskjp
Thank you for the interesting work.
I have checked the SemnaticUSL dataset, and it seems that sequence 00 to 08 has some annotation, but they a lot of missing/unlabeled/ignored points.
i.e., ss = np.fromfile("00/labels/000001.label", dtype=np.int32)
np.unique(ss, return_counts=True)
(array([ 0, 44, 50, 70, 71], dtype=int32), array([125133, 474, 4297, 767, 401]))
Can you please tell me which sequence you have used for validating/Evaluating the model on this SemnaticUSL dataset?
Thank you.
Dear @maskjp
Thank you for your impressive work.
I would like to know which field ( range, intensity, reflectivity ..... ) did you for the dataset and where the values normalized in anyway as i am having trouble with the results after training a network with your dataset. ( Using the same Lidar ).
Also why is the field of view set to 16.6 when the OS1-64 can do 22.5 ?
Eager to hear from you soon.
Regards,
Rohith.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.