2020 IEEE GRSS Data Fusion Contest

Organized by yky - Current server time: Nov. 30, 2020, 6:03 p.m. UTC

Previous

Track 2
March 6, 2020, noon UTC

Current

Future Development Phase
March 20, 2020, noon UTC

End

Competition Ends
Never

Global Land Cover Mapping with Weak Supervision

The 2020 Data Fusion Contest, organized by the Image Analysis and Data Fusion Technical Committee (IADF TC) of the IEEE Geoscience and Remote Sensing Society (GRSS) and the Technical University of Munich, aims to promote research in large-scale land cover mapping from globally available multimodal satellite data.

The task is to train a machine learning model for global land cover mapping based on weakly annotated samples. The 2020 IEEE GRSS Data Fusion Contest consists of two challenge tracks:

  • Track 1: Land cover classification with low-resolution labels
  • Track 2: Land cover classification with low- and high-resolution labels

For details about the 2020 IEEE GRSS Data Fusion Contest: http://www.grss-ieee.org/community/technical-committees/data-fusion/

Evaluation

The problem is to predict land cover classification maps. Each input is a triplet of Sentinel-1 dual-pol SAR, Sentinel-2 multispectral, and MODIS-derived land cover images. You are intended to submit 2D land cover classification images in TIFF with the same height and width (i.e., 256x256 pixels) as the input image.

The contest consists of three phases:

  • Phase 1 (Development phase): We provide you with the SEN12MS dataset for training and additional validation images (without any corresponding high-resolution labels) from December 13, 2019. You can submit prediction results for the validation set to Codalab to get feedback on the performance from January 13 to March 1, 2020. The performance of your BEST submission will be displayed on the leaderboard. In parallel, you need to submit a short paper of 1-2 pages clarifying the used approach, the team members, their Codalab accounts, and one Codalab account to be used for Tracks 1 and 2 by February 28, 2020 to be eligible to enter Phases 2 and 3. Please send a paper to iadf_chairs@grss-ieee.org using IGARSS paper template.
  • Phase 2 (Track 1): The test data set (without any corresponding high-resolution labels) will be released on March 1, 2020. You can submit your predictions of land cover maps from March 1 to March 6, 2020.
  • Phase 3 (Track 2): We provide you with semi-manually generated high-resolution labels for the validation set. You can submit your predictions of land cover maps from March 6 to March 20, 2020. In parallel, you need to submit a short paper of 1-2 pages describing the approach used for Track 2 by March 25, 2020. Please send a paper to iadf_chairs@grss-ieee.org using IGARSS paper template. After evaluation of the results, we will announce seven winners from two tracks on March 27, 2020.

Performance is assessed using average accuracy (AA) of all classes.

The winners will have approximately one month (submission deadlline: April 24, 2020) to write their manuscript that will be included in the IGARSS 2020 proceedings. Manuscripts are 4-page IEEE-style formatted. Each manuscript describes the addressed problem, the proposed method, and the experimental results.

Terms and Conditions

Participants of this challenge acknowledge that they have read and agree to the following Contest Terms and Conditions:

  • The data can be used in scientific publications subject to approval by the IEEE GRSS Image Analysis and Data Fusion Technical Committee and by the Technical University of Munich on a case-by-case basis. To submit a scientific publication for approval, the publication shall be sent as an attachment to an e-mail addressed to iadf_chairs@grss-ieee.org.
  • In any scientific publication using the data, the data shall be referenced as follows: “[REF. NO.] 2020 IEEE GRSS Data Fusion Contest. Online: http://www.grss-ieee.org/community/technical-committees/data-fusion”.
  • Any scientific publication using the data shall include a section “Acknowledgement”. This section shall include the following sentence: “The authors would like to thank the research group for Signal Processing in Earth Observation at the Technical University of Munich for providing the data used in this study, and the IEEE GRSS Image Analysis and Data Fusion Technical Committee for organizing the Data Fusion Contest.
  • Any scientific publication using the data shall refer to the following papers:
    @Article{isprs-annals-IV-2-W7-153-2019,
       author = {Schmitt, M. and Hughes, L. H. and Qiu, C. and Zhu, X. X.},
       title = {SEN12MS – A curated dataset of georeferenced multi-spectral sentinel-1/2 imagery for deep learning and data fusion},
       journal = {ISPRS Annals of Photogrammetry, Remote Sensing and Spatial Information Sciences},
       volume = {IV-2/W7},
       pages = {153--160},
       year = {2019}
    }
    @Article{9028003,
       author = {Yokoya, N. and Ghamisi, P. and Haensch, R. and Schmitt, M.},
       title = {2020 IEEE GRSS Data Fusion Contest: Global Land Cover Mapping With Weak Supervision [Technical Committees]},
       journal = {IEEE Geoscience and Remote Sensing Magazine},
       volume = {8},
       number = {1},
       pages = {154--157},
       year = {2020}
    }

Development Phase

Start: Jan. 13, 2020, noon

Description: Development phase: tune your models and submit prediction results on validation data. Do not forget to submit the short paper about your methodology before February 28, 2020!

Track 1

Start: March 1, 2020, noon

Description: Track 1: tune your models with low-resolution labels and submit prediction results on test data.

Track 2

Start: March 6, 2020, noon

Description: Track 2: tune your models with low-resolution labels and a limited number of high-resolution labels and submit prediction results on test data.

Future Development Phase

Start: March 20, 2020, noon

Description: Future-development phase: tune your models and submit prediction results on test data.

Competition Ends

Never

You must be logged in to participate in competitions.

Sign In
# Username Score
1 WHU_YuXia 0.5747
2 cyx669521 0.5377
3 UW 0.5018