TII Uses Crowdsourcing Competition to Discover Machine Learning-Based Solutions

Dr. Najwa Aaraj, the Chief Researcher for Cryptography Research Center and Autonomous Robotics Research Center, at TII, shares more information about this competition

Tell us about the Autonomous Robotics Research Center.
The Autonomous Robotics Research Center (ARRC) at the Technology Innovation Institute is the region’s leading organization to advance robotics, computer vision, and bio-inspired technologies.

We help research, develop, and apply transformative technologies to shape innovative robotic solutions that are fully independent of human intervention and infrastructure while remaining capable of interacting with humans and other biological systems in the environment.

Our robotics specialization includes perception, communication, control and decision-making, bio-inspiration, and modular and self-organized autonomous systems. We explore their application across multiple environments: air, land, on-sea, and underwater.

What is the Infrared Tracking Challenge all about?
Autonomous ground vehicles can understand their environment by applying deep-learning-based algorithms on top of perceptual information acquired from their sensors. At night, however, due to the low light conditions, the use of a standard RGB imager is not adequate.

The Infrared Tracking Challenge brings together innovators, technical experts, research institutes, and university students from around the world to develop approaches to efficiently track objects in the dark, in both structured and unstructured environments – the objects in question could include pedestrians, buggies, and motorcyclists, to name a few.

TII is using this crowdsourcing competition to discover machine learning-based solutions that can detect, track and re-identify objects in the dark, with only thermal or heat data.

Who is the target audience for this challenge and how will the participants be evaluated?
The target audience for the Infrared Tracking Challenge is any researcher or innovator or university student or a research institute that can shape working algorithms to track objects. The Challenge follows stringent judging criteria. Participants will need to show a High Order metric for evaluating multi-object tracking. The top solution submission’s score will determine the curve against which the other solutions are scored, and a threshold will be set based on the number of quality submissions received to allow them to proceed to Stage 2.

This stage will have a re-id criterion (instance reidentification) and the participants scoring high here will have their points added to their Stage 1 scores and move to the final stage. In Stage 3, the solutions will be integrated into the AI-embedded board and evaluated for compatibility and inference time.

What do the participants stand to gain from this challenge?
Three winners will be chosen out of all the participants and will be awarded a total of $75,000. First-place winners will be awarded $40,000, while second and third-place will be awarded $20,000 and $15,000 respectively.

In addition to the prize money, participants gain valuable networking experience by working with a global intellectual community and helping contribute to building an R&D ecosystem that reinforces the status of Abu Dhabi and the UAE as a global hub for innovation.

Is the challenge only open to participants from the UAE?
TII’s Infrared Tracking Challenge is open to all innovators, startups, research institutes, and university students globally.

What role does automation play in industries today?
Together with new technologies like artificial intelligence (AI) and machine learning (ML), and robotic process automation (RPA), automation has found application within various industries as a solid business strategy. Some of the advantages of this application include increased productivity, higher rates of production, better quality service, and the efficient use of materials.

In the case of infrared thermal imaging and tracking, these technologies can be employed in sectors such as healthcare, industry, emergency services, or disaster management. In healthcare, thermal imaging can reveal inflamed tissue, detect muscle injuries or diabetes, monitor peripheral circulation disorders, and identify tumors.

Industry professionals can use thermal imaging technology to analyze or keep a close eye on any sensitive components in a petrochemical facility or high-risk installation subject to Seveso legislation. Furthermore, it can ensure process control in an in-service oven or track tank levels in the petrochemical industry to help detect and prevent any integrity defects that may cause a serious fire or another catastrophic event.

Thermal imaging can also be used in the event of an emergency or disaster – in firefighting, for example. By rendering infrared radiation as visible light, this technology allows firefighters to see areas of heat through smoke, darkness, or heat-permeable barriers.