Outcome Announcement

The ISPRS Individual Tree Crown Segmentation (ITC) from High Resolution Images Contest was closed at UTC+0 00:00 on 22.06.2024. A total of over 200 participants from more than 40 teams out of 13 countries participated the contest. Congratulations to all participants who had successfully accomplished all the assignments of the contest! Your contributions to the contest and to the overall development of this research field are greatly appreciated!

The main objective of this contest is to reveal the state of the art of the methodological development for image-based ITC, and to clarify the remaining barriers and challenges to guide further explorations in the field. Therefore, the ranking and awarding aim to encourage continuous studies, instead of a rigid judgment on model performances. Other relevant contests are expected in the coming years and we hope all participants keep your spirits and interests in our coming contests.  

The ranking of submitted outcomes was performed on the web-based system in real time, which can be followed by all participants at the same time as the organizers. The models of the top six teams were validated and their outcomes were replicated before this announcement. Below is a complete list of all participants from the top six teams according to the evaluation of the submitted ITC results during the final test stage.

The official ceremony will be held in Nov., 2024 in ISPRS Technical Commission III Mid-term Symposium “Beyond the canopy: technology and applications of remote sensing”, Belém, Pará, Brazi. All participants are welcomed to join the ceremony.

Congratulations to the teams!


Rank 1: Team casmli

Ruiqian Zhang

Institute of Photogrammetry and Remote Sensing, Chinese Academy of Surveying and Mapping

Yuxing Xie

Research Institute of Urbanization and Urban Safety, University of Science and Technology Beijing

Xiaogang Ning

Institute of Photogrammetry and Remote Sensing, Chinese Academy of Surveying and Mapping

Hanchao Zhang

Institute of Photogrammetry and Remote Sensing, Chinese Academy of Surveying and Mapping

Rank 2: Team XXXR

Baocheng Dou

Zhejiang Lab

Xian Zhang

China Areo Geophysical Survey & Remote Sensing Center for Natural Resources

Shengbiao Wu

Future Urbanity & Sustainable Environment (FUSE) Lab, Division of Landscape Architecture, Department of Architecture, Faculty of Architecture, The University of Hong Kong

Xingwen Lin

Zhejiang Normal University

Dongqin You

Aerospace Information Research Institute, Chinese Academy of Sciences

Rank 3: Team DDD

Lingyun Zhang

Guangzhou Kugou Computer Technology Co., Ltd

Rank 4: Team flytocc

Yang Nie

CERI Digital Technology (Beijing) Co. Ltd. 

Xiaoying Nie

Zhejiang University

Rank5: Team mikky

Yan Miao

School of Ocean Engineering and Technology, Sun Yat-sen University

Feng Liu

School of Ocean Engineering and Technology, Sun Yat-sen University

Hu Chen

Xi’an University of Technology

Xu Guo

School of Ocean Engineering and Technology,  Sun Yat-sen University

Jinxin Shao

School of Ocean Engineering and Technology, Sun Yat-sen University

Rank 6: Team yuehhhh

Yinlei Yue

China University of Geosciences (Wuhan)

Jia Liu

China University of Geosciences (Wuhan)

Chongjiu Deng

China University of Geosciences (Wuhan)

Qi Xiong

China University of Geosciences (Wuhan)

Shiheng Fang

China University of Geosciences (Wuhan)

The final stage of the ISPRS ITC Segmentation Contest started. Scores/results are being checked.

The deadline of information and material submission: 25th Jun.

The phase 2 of the ISPRS ITC Segmentation Contest has finished. The results is being checked and confirmed.

Please note:

  • Every participant registered in the contest,regardless of the final ranking, is expected to submit the team information, thus, the participation record can be prepared.
  • Every participant who have submitted results successfully and received a valid rank in Codabenchis expected to submit a 1-2 page Word file of the description of your ITC segmentation method.
  • In case any of the original top 6 teamsfailed to submit the required information, the next best ranking team(s) who submitted all the necessary information and materials will be promoted to the top 6 teams to replace the failed ones.
  • One contribution prizewill be rewarded to a team who has significantly contributed to the success of the contest. Therefore, please make sure you have submitted all the necessary information and materials.
  • The final deadline of submitting the necessary information and materialthat mentioned before will be 25th Jun. Or, the score/result of will be invalid.
  • The top 6 teams on the leaderboardin Condabench will be contacted by organizers of ISPRS ITC Segmentation Contest via email after 25th Jun in 2 days. Then, the top 6 teams will be required to submit dockers and codes to organizers for score/results check. Please pay attention to your email.
  • The winners are planned tobe announced by 15th  The ceremony will be held in Nov., 2024 in ISPRS Technical Commission III Mid-term Symposium “Beyond the canopy: technology and applications of remote sensing”, Belém, Pará, Brazi. At least one person of the team in the winning list is required to present at the ceremony to represent the team.

Please submit all required information and materials via email to: And, the organizers will contact the top 6 teams by this email as well.

Team's Information Submission

Email Address for Submission:

For the management purpose, please send the information of your team by email:

  • please send the real name, affiliation, position(staff/under-graduate student/master student/Ph.D student) of each member of your team; there are maximum 5 members in one team;
  • list the namesaccording to their contributions, i.e., the highest goes the first;
  • provide the team name in Englishdisplayed on the platform of the contest in codabench.

Please sent the information by email to before June 21st, 2024. The information of your team will be use for the awarding and statistics. Please ensure the authenticity of the information.

The Testing Phase Starts

Deadline of Phase_2-Testing_phase: UTC + 0 00:00 June 22nd

The testing phase of the ISPRS international contest of individual tree crown (ITC) instance segmentation starts.

Please note:

  • There are 9 testing sets collected from different areas around the world released. The resolution of 2 new released dataset, Dataset_10 and Dataset_11, is 2 cm.
  • Only ONE model is allowed to be used for the ITC mask inference of all images of the final testing stage.
  • The performances of the participants will be evaluated based on the ranksof average ap50 in the final testing stage.
  • The top 6 teams are required tosubmit the dockers of their models to the organizer to confirm the results. The organizers will contact these teams via email in this regard. In case any participant(s) within the top 6 teams cannot be reached after three email notifications, or the provided model in docker cannot resemble the online reported outcomes, or the submitted docker does not work properly, the corresponding team will be removed from the top 6 list, and the next best ranked team(s) will be promoted to the top 6 teams. The final top 6 teams will be announced.
  • All participants, regardless their ranking, should senda Word file to describe their models after the testing phase by email.

The final testing stage closes at UTC+0 00:00 on June 22nd (UTC+1 01:00 on June 22nd in Paris; UTC+8 08:00 on June 22nd in Beijing; UTC-7 17:00 on June 21st in Los Angeles; UTC-12 12:00 on June 21st), and the competition phase of this contest ends at the same time. No further modification on models or results is permitted.

Figure. The examples of the testing sets in phase 2


Dataset Update

Deadline of Phase_1-Validation_phase: UTC + 0 00:00 June 15th

The ISPRS international contest of individual tree crown (ITC) instance segmentation releases 4 new training sets:

  • Dataset_6: 1234 images with resolution of 3 cm;
  • Dataset_7: 184 images with resolution of 5 cm;
  • Dataset_8: 300 images with resolution of 10 cm;
  • Dataset_9: 206 images with resolution of 10 cm.

The ISPRS International Contest of ITC segmentation has released up to now 9 training sets, 1 testing set, and 1 sample code in the training phase. The training sets include 8,000+ images and 450,000+ ITC masks. The training data are collected around world by multiple platforms, and cover various forest types and scenarios, e.g., boreal forest, subtropical forest, tropical forest, urban forest, etc.

Figure. The examples of all of the training sets in phase 1


Updates: Dataset, Sample Code, Schedule

New deadline: June 21st, 2024

The ISPRS international contest of individual tree crown (ITC) instance segmentation updates its datasets and schedule, and released a new piece of sample code as a reference to develop new segmentation method.

Schedule Update :

  • Deadline of validation phase: 00:00 UTC+0, June15th, 2024 
  • Deadline of testing phase: 00:00 UTC+0, June 22nd, 2024
  • Winner announcement: before July15th, 2024

Dataset Update:

  • The datasetsDataset_1 and Dataset_2 have been updated and enriched.
  • Three new datasets are released: Dataset_3, Dataset_4, Dataset_5.
  • The sizesof all images are the same, i.e., 1024×1024 pixels.
  • Dataset_1:the resolution is 2 cm; the training set includes 1691 images.
  • Dataset_2: the resolution is 10 cm;the training set includes331 images.
  • Dataset_3: the resolution is 4.5 cm; the training setincludes 331 1200 images, and the validation set includes 275 images.
  • Dataset_4: the resolution is 10 cm; the training setincludes 400 images, and the validation set includes 100 images. 
  • Dataset_5: the resolution is 2 cm; the training setincludes 1721 images, and the validation set includes 441 images.
  • The validation_img_id.json is released to appoint each validation image with an ID. The participants should generate inference results based on the IDsfor precision evaluation.

Sample Code Release:

  • The sample code can be used to train and validate the ITC segmentation network and make inference as a reference for the participants to develop their own methods.
  • The sample code utilizes the Mask R-CNN for ITC segmentation. It inputs the PNG images and the annotation stored in MS COCO Format. The predicted masks of individual tree crown are stored in MS COCO Format.
  • The sample code was modified based on the detecron2, whichprovide a library for instance segmentation algorithm development based on the Mask R-CNN.
  • The participants may follow the to execute the program.

Figure. The examples of the training sets and the sample code


ISPRS ITC Segmentation Contest Launched

Jan 29th → May 24th, 2024

Canopies play an essential role in biophysical functions of forest environments. Hence, individual tree delineation (ITD) from remote sensing (RS) data has been becoming one of the most important topics in forest RS research during past two decades. The International Society of Photogrametry and Remote Sensing (ISPRS)’s international contest of individual tree crown (ITC) instance segmentation aims to clarify the state-of-the-art of the tree crown instance segmentation from high-resolution earth observation images through a benchmarking study of different methodological approaches.

Abundant dataset gather around world is provided for model development. Recently, there are about 2000 1024×1024 image were published as training set and validation set. Besides, fine-grained ITC delineation masks in MS COCO format were provided at the same time. More data will be updated constantly during the contest.

The ISPRS ITC Segmentation Contest has be launched at Jan 29th, and will be ended at May 24th. The teams win the first-place, secon-place, and third-place prize win get 3k, 2k, and 1k dollars respectively.

Looking forward to your participation!

If you have any question, please contact Xinlian Liang:



Maria 1

Maria 2

Indoor 2

Indoor 1



The International Society for Photogrammetry and Remote Sensing is a non-governmental organization devoted to the development of international cooperation for the advancement of photogrammetry and remote sensing and their applications. The Society operates without any discrimination on grounds of race, religion, nationality, or political philosophy.

Our Contact

Leibniz University Hannover
Institute of Photogrammetry and GeoInformation
Nienburger Str. 1
D-30167 Hannover