ACTIA continues to take part in the MOT Challenge (Multiple Object Tracking), the global benchmark competition for tracking objects in a video scene. After initially being ranked 8th in 2020, this time ACTIA is ranked 3rd worldwide for another challenge proposed by the MOT Challenge, reflecting the quality of the object tracker it has developed. This challenge is an opportunity for the ACTIA group to put its research to the test on the world stage, testing functions that will expand its services on offer in the automotive sector.
RESEARCH APPLIED TO THE AUTOMOTIVE MARKETS
The objective of the challenge is to follow the targets identified in a scene over time. The goal is to connect detections of the same object over time, even to fill in missed targets or remove any false positives. This function has many applications, such as monitoring and analysing the movement of people, animals and vehicles.
The MOT challenge is currently the main benchmark for the comparative analysis of algorithms for tracking multiple targets in a video scene. The challenge focuses on tracking pedestrians. The result rows calculate the average performance across all the sequences proposed by the challenge.
ACTIA is ranked 3rd in performance among the published approachesin the private detector category.
DETECTION OF HAZARDOUS SITUATIONS
Multiple-object tracking via video is particularly useful in the automotive sector for tracking the trajectories of cars/pedestrians/cyclists around the vehicle and therefore detecting hazardous situations. ACTIA sees applications for this:
- in the public transport sector, by closely monitoring the environment of city buses, especially when they stop
- and on construction sites where increasingly automated machines require the integration of analysis, modelling and AI functions.
The MOT Challenge is an opportunity for the ACTIA group to take part in a global public demonstration of multi-person tracking in a context of varied, crowded scenes, making for a particularly difficult environment to monitor.
ACTIA will publish the results of its work soon in a scientific journal.
So, watch this space…
