Giter Club home page Giter Club logo

kitti360labeltool's Introduction

KITTI-360 Annotation Tool

KITTI-360 Annotation Tool is a framework that developed based on python(cherrypy + jinja2 + sqlite3) as the server end and javascript + WebGL as the front end. It is the annotation tool used to annotate the KITTI-360 dataset. Please find details of annotation instruction here.

Dependences

sudo apt install python-pip
pip install jinja2
pip install cherrypy
sudo apt install sqlite3 libsqlite3-dev

Quick Demo

  • Download the data for the demo with the following command:
./download_demo_data.sh
  • Start the web server with the following command:
./run_demo.sh

Demo in Details

1. Setup the data

Prepare data and put it in public/data/, or you can create a symbolic link from where the data is to public/data/

2. Configure user-task assignment

The user and task data is kept in db_import/. Example files have been included in db_import/*.tmp.

  • users.txt

    Each line is one user information in the following order:

    Email Name UserId Password IsAdmin
    
  • taskLists.txt

    Each line is one user-task information in the following order:

    TaskId UserId Editable
    

    where Editable is a number indicating:

    Editable = 0: readonly (User can only view the annotation)

    Editable = 1: normal (User can view, annotate, and submit results)

    Editable = 2: playground (User can view, annotate, but not submit results)

Note: space is not allowed in each item. So use '_' or other characters to connect words

3. Setup database

Once the user/task file is setup based on the above rules, to set up database, run

python create_db.py

4. Setup the host and port address

Setup the host and port address in server.conf, here is an example:

[global]
server.socket_host = "127.0.0.1" # host to be modified
server.socket_port = 8080        # port to be modified (e.g. 2000)

5. Start the server

python labelApp.py

6. Load the web page

Type http://host:port in the browser to load the web page. For example in the demo the web app is started locally in http://127.0.0.1:8080.

7. Get annotation results

Annotation results are saved in xml file under public/results/. We provide util functions to parse the xml file in https://github.com/autonomousvision/kitti360Scripts/blob/master/kitti360scripts/helpers/annotation.py#L353.

Folder Structure

.(ROOT)
|
+-- assets
|   |
|   +-- css
|   |
|   +-- javascripts: javascript source code
|
+-- db: generated database (by running create_db.py)
|
+-- db_import
|    |
|    +-- user-task configuration files
|
+-- public
|    |
|    +-- backup: backup XML annotation files, which are automatically saved during annotation
|    |
|    +-- data: annotation data
|    |
|    +-- resource: resource files such as icon
|    |
|    +-- results: resultant XML annotation files
|    |
|    +-- mapping.txt: label mergining file
|    |
|    +-- colorList.txt: label color mapping file
|
+-- views: HTML pages

Common Problems and Solutions

Q1: "socket.error: (98, 'Address already in use')"

A1: Kill the existing processor and re-restart the server. To kill the existing processor:

  1. Run ps aux | grep python in command-line.

  2. Find the PID with user name = annot, and run sudo kill -9 PID.

Citing KITTI-360 Annotation Tool

If you find this code helpful in your research, please use the following BibTeX entry.

@article{Liao2021ARXIV, 
   title   = {{KITTI}-360: A Novel Dataset and Benchmarks for Urban Scene Understanding in 2D and 3D}, 
   author  = {Yiyi Liao and Jun Xie and Andreas Geiger}, 
   journal = {arXiv.org},
   volume  = {2109.13410},
   year    = {2021}, 
}

License


MIT License

Copyright (c) 2021 Autonomous Vision Group

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.


kitti360labeltool's People

Contributors

ch-sa avatar yiyiliao 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

kitti360labeltool's Issues

Keybindings seem different than described in paper

Hi @yiyiliao,

great that you published and open sourced your annotation tool here. I was just a little confused about the key bindings, as they seem differently configured than described in the paper.

grafik

The paper says the E key switches to scaling, while the R key enables rotation mode. Yet in the software it seems to be implemented vice versa.

Also it was interesting seeing your approach as I implemented a similar software last year for labeling 3D bounding boxes in point clouds called labelCloud. We also presented and published it in CAD21, maybe the corresponding publication is also of interest for you.

UPDATE:

  • I added a pull request that fixes the issue. It should be ready to merge.

Best regards,

Christoph

How to change the size of the bounding box

Many thanks for your efforts and this great tool. It looks to be very promising. Initially there's is a indent error on labeApp.py due to inconsistent space and tab usage.

Once I fix this locally I was able to launch the web app and login.

I am able to create a 3D bounding box using the label buttons, for example to label a car. I can move around the bounding box. However I don't see any clear way to resize the box to fit the car.

I tried various things like double clicking the inside of the box, the arrows, etc.

Can you please provide some guidance on this?

I wonder if your sample annotation mentioned in instruction still works ?

Thanks for your great work!

We notice that you have post a sample annotation webside as shown in the below figure. However, as I visit this link, it reports 502 bad gateway error. I wonder whether this website still works or not. Since the functions reported in your instruction really fits us, we can't wait to try the tool.

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.