Giter Site home page Giter Site logo

kaist-avelab / k-radar_v1 Goto Github PK

View Code? Open in Web Editor NEW
112.0 14.0 11.0 28.38 MB

4D Radar Object Detection Dataset and Benchmark for Autonomous Driving in Various Weather Conditions

Python 90.72% C++ 2.27% C 0.22% Cuda 0.68% CMake 0.51% MATLAB 5.59%
3d-object-detection autonomous-driving 4d-radar 4d-radar-tensor adeverse-weather

k-radar_v1's People

Contributors

dongheepaek avatar kaist-avelab avatar ktirta avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

k-radar_v1's Issues

Permissions on GoogleDrive file meta_1.zip

Thank you for the wonderful dataset and offering Google Drive to download the data!

The file meta_1.zip (for sequence 1) does not have permissions for viewers to download it, unlike all other files. This needs to be manually changed :)

image

Radar pointcloud & Max range

Thanks for the great research!

I have a few questions.

  1. Does the dataset contain 4D radar pointcloud?
  2. What is the maximum distance annotated?

About specification for dataset and how to get development kits

Hi, Thank you for building and sharing this dataset. It is very useful. Great job!
Could you please provide a specification for how to use dataset, like data format, time synchronization and so on? I just downloaded part of zips (It is too big to be downloaded completely) but didn't found. Maybe it is in one of them?
And, the project mentioned that all development kits and labeling tools will be released too. Could you also show me how to get them please?
Thanks again!

cannot use download link

I tried to download the dataset,unfortunately i was not able to download it.(screenshot attached
k radar dataset
)

The NAS download is still unavailable

Hello, thanks for your work on releasing the dataset.
However, I still cannot log in to your NAS and download any data with your link.
Could you please fix it and make your work really available to us?
Thanks again for your excellent work.

That's what I see after click your link:

Screenshot from 2022-12-19 16-20-53

Issues about data format

Hi! Thanks again for sharing the great job!
I want to know is the complex number format stored in the 4DRT(4D Radar tensor) data? And when will the dataset be made public?

Opening the zipped radar data

Thank you for the wonderful dataset and sharing Google Drive links to download it!

I have noticed there are no instructions on how to unzip/merge the three archives for radar data in each sequence. Take for example sequence 2, there are files radar_02.z01, radar_02.z02 and radar_02.zip. The final of those three contains reference to all 628 *.mat files contained in the multi-part archive (seen using unzip -l radar_02.zip). However, if extracted using unzip, only 148 files are obtained (with some warnings about missing data). It seems like it skips radar_02.z01 and radar_02.z02.

I have tried concatenating the three zip files and then opening a single file, but that also fails:

$ cat radar_02.z01 radar_02.z02 radar_02.zip > radar_combined_z01_z02_zip.zip
$ unzip radar_combined.zip

Archive:  radar_combined_z01_z02_zip.zip
error: End-of-centdir-64 signature not where expected (prepended bytes?)
  (attempting to process anyway)
warning [radar_combined_z01_z02_zip.zip]:  zipfile claims to be last disk of a multi-part archive;
  attempting to process anyway, assuming all parts have been concatenated
  together in order.  Expect "errors" and warnings...true multi-part support
  doesn't exist yet (coming soon).
warning [radar_combined_z01_z02_zip.zip]:  27917287424 extra bytes at beginning or within zipfile
  (attempting to process anyway)
file #1:  bad zipfile offset (local header sig):  27917287428
  (attempting to re-compensate)
error: invalid zip file with overlapped components (possible zip bomb)

Instructions found in the K-Lanes dataset unfortunately do not work in this case either:

To unzip the file in linux system(eg. Ubuntu 20.04)

// sudo apt-get install default-jdk
// jar xvf (file_you_downloaded).zip

This gives errors (for individual parts and the concatenated file from above) such as:

 $ ▶ jar xfv radar_combined_z01_z02_zip.zip
java.util.zip.ZipException: invalid CEN header (bad signature)
        at java.base/java.util.zip.ZipFile$Source.zerror(ZipFile.java:1607)
        at java.base/java.util.zip.ZipFile$Source.initCEN(ZipFile.java:1557)
        at java.base/java.util.zip.ZipFile$Source.<init>(ZipFile.java:1308)
        at java.base/java.util.zip.ZipFile$Source.get(ZipFile.java:1271)
        at java.base/java.util.zip.ZipFile$CleanableResource.<init>(ZipFile.java:733)
        at java.base/java.util.zip.ZipFile$CleanableResource.get(ZipFile.java:850)
        at java.base/java.util.zip.ZipFile.<init>(ZipFile.java:248)
        at java.base/java.util.zip.ZipFile.<init>(ZipFile.java:177)
        at java.base/java.util.zip.ZipFile.<init>(ZipFile.java:148)
        at jdk.jartool/sun.tools.jar.Main.extract(Main.java:1388)
        at jdk.jartool/sun.tools.jar.Main.run(Main.java:410)
        at jdk.jartool/sun.tools.jar.Main.main(Main.java:1680)

Any help would be greatly appreciated :)

Total dataset size

Hello,

I can now access the dataset link. It has 54 zip files with each having ~200 GB+ data.

Does that mean the total size of the dataset is approx. 50 x 200 = 10000 GB+ ???

Could you please clarify this?

Thanks,
Kevin

Download via NAS server is slow - download via terminal not possible

After retrying many times, I was able to connect to the NAS server and view the files.

There are ~50 zip archives, each ~200 GiB in size. I reached a steady download speed that meant > 3 weeks for a single zip file. The download can only take place via the browser itself, so successfully finishing a download over 3 weeks will be difficult (I haven't managed it).

Would it be possible to make links that can be downloaded in a terminal e.g. using wget or curl? Alternatively, making files available via something like AWS or GCS would make the data much more accessible; higher download speeds and terminal CLI tools for headless downloads.

Pretrained models

Thanks for publishing your interesting work!
I wonder if you plan to share your some pretrained model?, so we can try the model without training from scratch. As the dataset is huge to download.
Thanks.

DREA to YXZ,the detail about the doppler processing

Hi, I have read the Matlab code about the radar data processing. Here are my questions?
1、when transfer the DREA to XYZ, the code just mean the frist 10 doppler array? I think because the objects is dynamic, why not mean all the doppler dimension?
2、the speed is ranging from the -1.9 to 1.9m/s, and I think the radar form may be designed to detect much larger speed by using different chirps?

Hope to know more detail about the cfg of the radar.

IMU and GNSS data

Thanks for the contribution
I'd like to know if the raw IMU and GNSS data is available?
can this dataset be used to slam?

ADC data

I am very interested in your work. I downloaded the seq_1 data, but I did not find the original ADC data, nor the 4DRT data. Could you give me a hand !
my fold list is
image
has not find
image

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.