## EO-IR Object detection challenge

Greeting to all the participants, this is the first edition of the challenge hosted by **Artificial Intelligence and Robotics laboratory (AIRL) from Indian Institute of Science (IISc), Bengaluru**. You are required to train object detection model using aerial Electro-Optical (EO) images (optionally can use Infrared (IR) images provided with IndraEye dataset) and obtain best predictions on EO and IR test data. This setup is to use EO and IR modalities effectively to generalize object detection models on both domain and varying slant angles without forgetting also adapt from one domain to the other domain effectively. 

# 🔥 Final Winners:
First place 🥇: **Team CDA**
Second place 🥈: **Vatsal Maheshkumar**
Third place 🥉: **Bug Smashers**


# ‼️Important update to all participants:
Dear participants,
As the submissions for the competition is closed, we request all the teams to submit the checkpoint, validation code and environment yaml file for us to evaluate and announce the winners. Please fill the form below by attaching the appropriate file.

https://forms.gle/ZfwJeC8aGkav7AqY7
We will be closing the form on 15th July 2025 (11:59 pm), please try to complete it as soon as possible.


## What’s New in This Challenge?
Unlike other simple object detection challenges, this challenge enables participants to work on real-world data i.e, IndraEye which is the only available EO-IR object detection dataset collected using drone in the city of Bengaluru. Achieving better performance on both EO and IR domain enables the use of model in various applications such as traffic management, defense applications, and much more. 


## How the challenge is important for the participancts?
**Win prizes**: Certificates and Shoutouts on platforms like LinkedIn and Hugging Face with a chance to feature on our competition blog. 
**Paper submission**: Top two participants will be called to present the work as a paper to the International Conference on Emerging Technology in Autonomous Aerial Vehicles (ETTAV) conference held in IISc Bengaluru (Detail of conference: [link](https://sites.google.com/view/etaav2025)). 
**Boost your portfolio**: Showcase your AI & ML skills in real-world object detection.


If you aim to publish your work using this dataset, kindly consider citing the Paper](https://arxiv.org/abs/2504.15728) using:
@misc{d2025sagasemanticawaregraycolor,
      title={SAGA: Semantic-Aware Gray color Augmentation for Visible-to-Thermal Domain Adaptation across Multi-View Drone and Ground-Based Vision Systems}, 
      author={Manjunath D and Aniruddh Sikdar and Prajwal Gurunath and Sumanth Udupa and Suresh Sundaram},
      year={2025},
      eprint={2504.15728},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2504.15728}, 
}

This competition is supported by SwaYaan, Ministry of Electronics and Information Technology (MEITy)