Welcome to the 9th ABC Challenge
Brought to you by Université du Québec à Chicoutimi (UQAC) in conjunction with ABC

Overview

Falls are a major concern in elderly care and assisted-living environments. Automatic fall-detection systems must not only distinguish falls from normal activities but must also operate continuously without generating excessive false alarms during everyday behavior.

The 9th ABC Challenge 2027 focuses on privacy-preserving fall detection using Ultra-Wideband (UWB) radar signals collected in a realistic smart-apartment environment.

Unlike conventional sequence-level classification tasks, participants receive a continuous, non-segmented radar data stream containing a mixture of Activities of Daily Living (ADLs) and fall events. Systems must automatically detect each fall and estimate its temporal boundaries.