FishTrackingChallenge2024

@International Symposium of Hierarchical Bio-Navigation 2024

We are hosting a multi-object tracking competition, Fish Tracking Challenge 2024, using the SweetFish dataset at the International Symposium of Hierarchical Bio-Navigation 2024!(https://bio-navigation.jp/navisympo2024/)

The task is very simple: tracking 10 fishes’ locations from the SweetFish dataset.

The SweetFish dataset captures the complex collective behaviors of aquatic animals, with a focus on sweetfish navigation. While understanding these patterns is crucial for various scientific disciplines, accurately tracking sweetfish, especially in groups, presents challenges. By developing an advanced tracking model, researchers can uncover the intricacies of aquatic movement and significantly advance this field. Dive in and make waves in the world of aquatic animal research!

Competition Start

2023-10-15 (AOE)

Competition End

2024-01-15 (AOE)

Task

Tracking 10 fishes’ locations (multi-object tracking)

Prize

Travel Fee Support

Contents:

Overview

Background

Collective animal behaviors are teeming with life and intricate behavioral patterns. The sweetfish, included in this dataset, offers a unique window into understanding animal navigation in water. For ethologists, ecologists, and mathematical and theoretical biologists, decoding these patterns is pivotal, but automatic tracking the navigation of sweetfish, especially when in schools, introduces a plethora of challenges.

Challenge

You can develop an automatic tracking model that can pinpoint the navigational patterns of sweetfish with the Sweetfish dataset including locations of 10 fishes. The dataset includes videos and bounding boxes of fishes. To analyze the behaviors recorded by videos, automatic detection and tracking of bounding boxes is necessary. We have train, development, and test datasets. By training your object detection and tracking algorithms using training dataset, you can estimate and submit the bounding box data in development and test datasets.

Video

Prize

Travel Fee Support

At least one person from the top-3 team was encouraged to attend the award ceremony on-site or online. We will support the travel fee (within Japan) for the attendance of up to 1 member from each winning team, who lives in Japan.

Winners Obligations

As a condition to being awarded a Prize, a top-5 winner must fulfill the following obligations. The detailed instructions will be sent to top-5 winners after the final submission deadline.

  • Submit your code so that we can check for cheating.
  • Submit a short report paper that describes the award methodology.

Detail

Data

The Dataset for this challenge is SweatFish dataset (you can download from [Google drive Link] ). You can use all dataset listed below:

Evaluation

The goal of this challenge is accurate tracking of 10 sweetfishes. Performance of your model will be evaluated based on HOTA (Higher Order Tracking Accuracy) score, which is a holistic and popular score in multi-object tracking (MOT). HOTA is designed to overcome many of the limitations of previous metrics. For example, you can see this web page for understanding.

Submission Format

To submit your results to this competition you must construct a submission zip file containing a single file named answer.txt. Sample submission files for Development Phase are available as follows:( submission.zip )

Rules for Fish Tracking Challenge 2024

A. Registration Required

Please submit a team registration form before you join this competition even if you are a solo challenger. This is required to get prizes. There are no limitations for the team size. However, if you are creating multiple teams within the same lab, be careful not to violate the rules of PRIVATE SHARING.

When you want to merge your team with another team, please contact the Fish Tracking Challenge Admin Team via email.

B. No Private Sharing Outside Teams

Privately sharing code or data outside of teams is not permitted. It’s OK to share code if made available to all participants on the forum.

C. External Dataset

You may use data other than the competition data to develop and test your submission. However, you will ensure the external data is publicly available to everyone without any cost.

Results

Thank you to everyone who participated, we had 16 teams and 31 registered users on colab. We received a total of 163 submissions during the period. (As of January 16, 2024)

The top three teams in the competition are listed below. Congratulations! Please see this table ( Google Sheet) here for detailed scores.

1st Place

ytachioka (HOTA score=0.49)

Tachioka yuuki. (Denso IT laboratory)

2nd Place

simon787 (UWIPL) (HOTA score=0.47)

Hao Wang et al.(University of Washington Information Processing Lab)

3rd Place

maburto(LabTrack) (HOTA score=0.44)

Andres Mohali et al.(NAIST)

All Results

Rank Team HOTA score
1 ytachioka 0.49
2 simon787 0.47
3 maburto 0.44
4 _640_ 0.39
5 michaelibrahim 0.38
6 tomoon 0.35
7 xmba15 0.18

Awards Ceremony @International Symposium of Hierarchical Bio-Navigation 2024

The awards ceremony was held on March 11, 2024 at International Symposium of Hierarchical Bio-Navigation 2024. The top 1 and 3 winners gathered to discuss their solutions in a poster session. Congratulations again!

Picture with Top 1&3 Winners @International Symposium of Hierarchical Bio-Navigation 2024

Registration and Submission

Before submitting your prediction, registration is required. You need to register your information to both (1) CodaLab and (2) Google Form. Please follow the procedures below.

  • Solo challengers are also welcome! But don't forget to register as a solo team!
  • No limitation to team size. (But be careful about violation of the NO PRIVATE SHARING policy when you make multiple teams.)

Steps for Registration

  1. [Every Members] Make your codalab accounts. All members should have thier own accounts.
  2. [Every Members] Register to this challenge from Participate tab in the codalab page.
  3. [Team Leader] Submit the Google Form for team registration.
  4. (Wait for a while... Admin team will check and aprove your team.)
  5. You got email from codalab about approval and you can submit your prediction!

Submission & Leadersboard

Timeline

2023-10-01
Registration Open
Google form and competition page on codalab will be public.
2023-10-01
Registration Open
Google form and competition page on codalab will be public.
2023-10-15
Launch
You can submit your estimates to the competition site.
2023-10-15
Launch
You can submit your estimates to the competition site.
2024-01-10
Entry and Team Merger Deadline
If you want to join this competition or change team members, you must register until this date.
2024-01-10
Entry and Team Merger Deadline
If you want to join this competition or change team members, you must register until this date.
2024-01-15
Final Submission Deadline
You must submit your best results until the end of this day.
2024-01-15
Final Submission Deadline
You must submit your best results until the end of this day.
2024-01-31
Report Submission Deadline
Top 3 winners must submit a report of your solution to get the prize! Deadlines are subject to change.
2024-01-31
Report Submission Deadline
Top 3 winners must submit a report of your solution to get the prize! Deadlines are subject to change.
2024-03-11
Award Ceremony in the Internationla symposium
Top 3 awards will be presented
2024-03-11
Award Ceremony in the Internationla symposium
Top 3 awards will be presented

Terms of Use (License)

The SweetFish dataset consisting of video and location data will continue to be provided under "CC BY-NC-SA 4.0".

Cite

Takayuki Niizato, Kotaro Sakamoto, Yoh-ichi Mototake, Hisashi Murakami, Takenori Tomaru, “Information structure of heterogeneous criticality in a fish school”. bioRxiv 2024.02.18.578833; doi: https://doi.org/10.1101/2024.02.18.578833 . https://www.biorxiv.org/content/10.1101/2024.02.18.578833v1

Cite

Makoto M. Itoh, Qingrui Hu, Takayuki Niizato, Hiroaki Kawashima, Keisuke Fujii, “Fish Tracking Challenge 2024: A Multi-Object Tracking Competition with Sweetfish Schooling Data”.
arXiv preprint arXiv:2409.00339. https://doi.org/10.48550/arXiv.2409.00339
http://arxiv.org/abs/2409.00339

Github

Github tool

baseline program for tracking

Google colaboratory

Google colab tool

baseline program for tracking

Members

Keisuke Fujii
Nagoya Univ.

Makoto Itoh
Nagoya Univ.

Qingrui Hu
Nagoya Univ.

Adviser

Hiroaki Kawashima
University of Hyogo

Naoya Yoshimura
Osaka University

Atom Scott
Nagoya University

Data Provider

Takayuki Niizato
University of Tsukuba

Special Thanks

A

Alex Hoi-Hang Chan
University of Konstanz

Sponsors

Cynav: What is Hierarchical Bio-Navigation?