Leaf Segmentation Challenge (LSC)

To advance the state of the art in leaf segmentation and to demonstrate the difficulty of segmenting all leaves in an image of plants, we organize the Leaf Segmentation Challenge (LSC). For the challenge we release a training set (which contains raw images and annotations). Few days before the paper due date, a testing set will be released. Papers will be evaluated and ranked according to their outcome, the validity of the algorithm, and suitability of the approach. Only fully automated approaches will be accepted. Accepted papers will be presented either orally or in a poster session, and will appear in the proceedings. Should a large, high quality, number of papers be received, a collation study, summarizing algorithms and results, will be presented instead with authors presenting details in a poster session.
A jointly authored paper presenting the findings of the collation study may be invited to a high impact journal in computer vision (to be announced at the workshop). The collation paper will be compiled only from participants presenting at the workshop.


How to participate:

This challenge expired. In case you want to run tests on this challenge, please do let us know 

  1. Please read first the challenge terms and conditions
  2. Please register for the challenge by filling in the online registration form (link no longer available, please visit our dataset page). This registration is a mandatory step before downloading data and submitting results to the challenge.
  3. Upon reception of your registration form, you will receive a link to download the training dataset (as a single zip file), collected in our laboratories of top-view images of rosette plants. In particular, we share images of tobacco plants and arabidopsis plants. Tobacco images were collected using a camera which contained in its field of view a single plant. Arabidopsis images were collected using a camera with a larger field of view encompassing many plants, which were cropped. The images released are either from mutants or wild types and have been taken in a span of several days. Plant images are encoded as tiff files.
    All images were hand labelled to obtain ground truth masks for each leaf in the scene. These masks are image files encoded in PNG where each segmented leaf is identified with a unique integer value, starting from 1, where 0 is background.
    The archive contains an evaluation function (in MATLAB) for comparing segmentation outcomes between ground truth and algorithm results.
  4. Two days before the submission deadline registered authors will receive instructions on receiving the testing dataset and submitting testing set results to the challenge organizer.
  5. The organizers will evaluate the results, and outcomes will be sent back to the authors for inclusion in their submission.

Challenge Terms and Conditions:

  • All the data made available for the CVPPP 2014 Leaf Segmentation Challenge (LSC) can only be used to generate a submission for this challenge.
  • Results submitted to CVPPP 2014, can be published (as seen appropriate by the organizers) through different media including this website and journal publications.
  • By submitting an entry to CVPPP 2014 LSC, each team agrees to have at least a single member register to the accompanying workshop (held on September 12, 2014).

These guidelines follow those established by challenges in biomedical image analysis such as example 1 and example 2.


Deadlines for the Challenge 

Challenge opens for registration:         March 2014
Training data ready for download: April 2014
Testing data ready for download: June 9 2014
Submit results on testing data             June 11 2014, 11:59PM Pacific Time
Evaluation of results on testing data    June 12 2014
Submission deadline: June 15 2014, 11:59PM Pacific Time
Notification of acceptance: July 15 2014
Camera-ready paper: July 25 2014, 11:59PM Pacific Time

September 12 2014


Registration stays open till June 8 2014.


download flyer


Download LSC example images (8.6MB)

Phidas logo


EU and FP7 Logos




Some example images from data set