Falls constitute a leading cause of fatal and non-fatal injuries among older adults. Identifying individuals at high risk of falling through fall risk assessment is a crucial step in implementing effective fall prevention strategies. Exiting studies predominantly rely on wearable sensors and depth cameras for fall risk assessment, which can be either inconvenient or costly. To address these limitations, this preliminary study aimed to investigate the feasibility of distinguishing between older fallers and non-fallers based solely on data from a single RGB camera through a simple walking test. We employed an open-source human pose estimation algorithm (YOLOv8) to conduct gait analysis from a frontal perspective. A validation experiment involving 60 community-dwelling older adults (aged 65 years and older) revealed significant or at least marginally significant differences between faller and non-faller groups in the majority of spatiotemporal gait parameters (e.g., step time, step length, step velocity, stride time, stride length, stride velocity and cadence). These findings suggest that RGB camera-based systems hold substantial promise for older adults’ fall risk assessment. Such systems could be readily deployed in home-based settings, making fall risk assessment accessible and regular for a broader aging population.