Giter Club home page Giter Club logo

visdrone2018-det-toolkit's Introduction

VisDrone2018-DET Toolkit for Object Detection in Images


This is the documentation of the VisDrone2018 competitions development kit for object detection in images (DET) challenge.

This code library is for research purpose only, which is modified based on the PASCAL VOC [1] and MS-COCO [2] platforms.

The code is tested on the Windows 10 and macOS Sierra 10.12.6 systems, with the Matlab 2013a/2014b/2016b/2017b platforms.

If you have any questions, please contact us (email:[email protected]).


If you use our toolkit or dataset, please cite our paper as follows:


title={Vision Meets Drones: A Challenge},

author={Zhu, Pengfei and Wen, Longyin and Bian, Xiao and Haibin, Ling and Hu, Qinghua},

journal={arXiv preprint:1804.07437},

year={2018} }


For DET competition, there are three sets of data and labels: training data, validation data, and test-challenge data. There is no overlap between the three sets.

                                                        Number of images
  Dataset                            Training              Validation            Test-Challenge
  Object detection in images       6,471 images            548 images             1,580 images

The challenge requires a participating algorithm to locate the target bounding boxes in each image. The objects to be detected are of various types including pedestrians, cars, buses, and trucks. We manually annotate the bounding boxes of different objects and ignored regiones in each image. Annotations on the training and validation sets are publicly available.

The link for downloading the data can be obtained by registering for the challenge at Evaluation Routines

The notes for the folders:

evalDET.m is the main function used to evaluate your detector -please modify the dataset path and result path -use "isImgDisplay" to display the groundtruth and detections

DET Submission Format

Submission of the results will consist of TXT files with one line per predicted object.It looks as follows:


    Name                                                  Description
 <bbox_left>	     The x coordinate of the top-left corner of the predicted bounding box

 <bbox_top>	     The y coordinate of the top-left corner of the predicted object bounding box

 <bbox_width>	     The width in pixels of the predicted object bounding box

<bbox_height>	     The height in pixels of the predicted object bounding box

   <score>	     The score in the DETECTION file indicates the confidence of the predicted bounding box enclosing 
                     an object instance.
                     The score in GROUNDTRUTH file is set to 1 or 0. 1 indicates the bounding box is considered in evaluation, 
                     while 0 indicates the bounding box will be ignored.
<object_category>    The object category indicates the type of annotated object, (i.e., ignored regions(0), pedestrian(1), 
                     people(2), bicycle(3), car(4), van(5), truck(6), tricycle(7), awning-tricycle(8), bus(9), motor(10), 
<truncation>	     The score in the DETECTION result file should be set to the constant -1.
                     The score in the GROUNDTRUTH file indicates the degree of object parts appears outside a frame 
                     (i.e., no truncation = 0 (truncation ratio 0%), and partial truncation = 1 (truncation ratio 1% ~ 50%)).
<occlusion>	     The score in the DETECTION file should be set to the constant -1.
                     The score in the GROUNDTRUTH file indicates the fraction of objects being occluded (i.e., no occlusion = 0 
                     (occlusion ratio 0%), partial occlusion = 1 (occlusion ratio 1% ~ 50%), and heavy occlusion = 2 
                     (occlusion ratio 50% ~ 100%)).

The detections in the ignored regions or labeled as "others" will be not considered in evaluation. The sample submission of the Faster-RCNN detector can be found in our website.


[1] E. Park, W. Liu, O. Russakovsky, J. Deng, F.-F. Li, and A. Berg, "Large Scale Visual Recognition Challenge 2017", [2] T. Lin, M. Maire, S. J. Belongie, J. Hays, P. Perona, D. Ramanan, P. Dollar, and C. L. Zitnick, "Microsoft COCO: common objects in context", in ECCV 2014.

Version History

1.0.2 - Apr 27, 2018

The nms function is included. The annotations in ignored regions are removed.

1.0.1 - Apr 25, 2018

The display function of groundtruth and detection results are included.

1.0.0 - Apr 19, 2018

Initial release.

visdrone2018-det-toolkit's People


visdrone avatar



Recommend Projects

  • React photo React

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

  • Vue.js photo Vue.js

    ๐Ÿ–– 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. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

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 โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.