Biological motion perception: an algorithm for creating realistic human point-light-walkers

What is it about?

Human actions, such as walking, captured solely by moving dots on the major joints (known as biological motion) is immediately and effortlessly perceived, and this perceptual ability has captured the interest of researchers for over 50 years. Stimulus generation used in research on this ability rely on video recordings, or motion capture and sophisticated equipment to record 3D coordinates of actors' limbs. Now, an innovative algorithm has been developed to offer an important contribution to this research. This algorithm generates visually realistic computer animations of human Point-Light-Walkers (PLWs) from any conceivable viewing perspective and distance. It allows researchers to manipulate stimulus parameters that control body structure, viewing perspective, and distance. In addition, other parameters adjust walking styles, and the relative phases between limbs, shoulder, and hip motions to simulate relaxed walk, a walk with forward thrust, trot or pace walking, or combinations thereof. The simulation incorporates a ground level constraint, resulting in a realistic vertical body sway important for realistic walking motions. This not only enhances the authenticity of the animation but also leads to a local visual directional invariant, an important characteristic in real walking. Previous algorithms were confined to specific viewpoints, displayed unnatural walking, and lacked this crucial visual direction invariant. An additional algorithm simulates accretion-deletion cues caused by self-occlusions during the gait cycle. The computer-animated PLWs generated by the present algorithm thus offer unprecedented control over stimulus parameters. This flexibility makes it an ideal tool for researchers exploring biological motion perception from various angles, spanning fields from fundamental perception studies to clinical research and beyond. In conclusion, this algorithm providing researchers with a powerful and flexible tool to investigate human perception of biological motion from diverse perspectives. Links to video demos are found in the PsyArXive-preprint (link below) on page 8, the preprint also contain the algorithm and a github link to the program code written in visual basic 5.

Featured Image

Photo by Marc Schulte on Unsplash

Photo by Marc Schulte on Unsplash

Why is it important?

An algorithm is presented that provide researchers with a powerful and flexible tool to investigate human perception of biological motion from diverse perspectives (here biological motion refer to the relative motions of a collection of dots that mark the mayor joints of a human walking).