Tempo

When movement can be traced, what does continuity look like?

2025 | Personal Research Project

Advisor · Liam Young

supported by SCI-Arc

Tempo explores how the continuity of human motion can be revealed through computational means. Using a physically-based particle system, the project traces movement as a flow of force. When velocity, acceleration, and directional change shape evolving fields and visual trails, it encourages viewers to follow, interpret, and sense the dynamics embedded in motion.

Intention

Tempo grew out of my curiosity about how motion carries meaning beyond visible form. When watching dance performances, I became particularly drawn to what happens between poses: the momentum that persists, the hesitation before a shift, and the subtle forces that stretch and reshape space as the body moves. These qualities are often felt intuitively, yet they are difficult to isolate or describe using conventional motion representations.

Rather than treating movement as a sequence of discrete positions, Tempo investigates motion as a continuous flow of force. I am interested in the in-betweens of motion, particularly how velocity, acceleration, and directional change can be understood as a representation of “trace” that persists over time and shape behavior, rather than simply describing where a body has been.

This project serves as an early research exploration into how human motion can act as input for a computational system. By translating motion capture data into force fields and particle behaviors, Tempo asks how computational systems might sense continuity, rhythm, and momentum, instead of reacting only to static poses or events.

Through this process, I explore particles as a medium for recording motion over time. Unlike geometry, particles allow force to accumulate, disperse, and transform more fluidly and make temporal continuity visible. Tempo positions motion not as something to be reproduced, but as something that can actively shape a dynamic system, revealing how movement itself can become computational material.

Research

Tempo investigates how motion can be interpreted as force, and how a particle medium can document continuity as flow rather than form.

How can motion be translated into a continuous force landscape?

What becomes visible when force is represented as flow rather than geometry?

The goal is not to animate an object, but to let movement shape the behavior of a dynamic system.

Motion study / motion continuity Motion study: directionality & rhythm
Motion study / motion extraction Motion study: motion extraction
Continuity traces / trail experiments Continuity traces: trails as time-based record

Why particles

Compared to geometry, particles can represent force as flow—accumulating, dispersing, and transforming over time—creating a more continuous trace of motion.

What “continuity” means here

Continuity is treated as temporal coherence—how momentum carries across frames—rather than a sequence of discrete poses.

Technical Approach

SYSTEM PIPELINE Motion capture → force vectors → distance fields → particle behavior → trails SENSING INTERPRETING SIMULATING / RENDERING Human Movement full-body performance energy · rhythm · continuity Motion Capture (Input) Rokoko suit → trajectories timing · joints · motion curves Motion Interpretation vectors of directionality momentum · continuity Force-to-Field Translation signed distance / force fields flow · propagation · influence Particle Behavior Houdini · VEX rules velocity · turbulence · trails Visual Output trails visualize continuity force becomes flow ROKOKO (INPUT) SDF / FIELDS HOUDINI VEX

1) Motion Capture as Force Input

Full-body gestures were captured using a Rokoko motion capture suit and processed beyond skeletal animation. Motion data was interpreted as vectors of directionality, momentum, and temporal continuity, describing how force flows through the body over time. These vectors form the energetic input that drives the system.

2) Force-to-Field Translation

Motion-derived vectors were mapped into signed distance fields that encode acceleration, flow, and spatial influence. Rather than discrete coordinates, movement generates a continuously evolving force landscape that shapes system behavior.

3) Particle Behavior as Material

A custom particle system responds to the force fields by adjusting velocity, turbulence, and trail behavior. Particles behave as if reading the dancer’s energy—swirling, clustering, stretching, or dissolving in response to expressive motion.

Development

mocap data animation
Iteration 1 — Motion cleanup & stable vectors

Goal: preserve continuity while reducing noise.

Result: usable velocity/acceleration signals for force mapping.

Mocap cleanup
Before / after smoothing
Vector extraction
Vector extraction tests
Iteration 2 — Field mapping & behavior calibration

Goal: translate force into spatial influence (distance fields).

Result: stable propagation + readable flow responses.

Field visualization
Field visualization
Iteration 3 — Trail logic for continuity

Goal: make temporal continuity legible through trails.

Result: length/curl/density respond to acceleration & direction change.

Trail study A
Original Trail
Trail study B
Trail variant A: Add directional control for each particle strength: 20%
Final trail logic
Trail variant B: Add directional control for each particle, strength: 80%

Outcomes

Tempo produced a computational system that visualizes motion as continuous force rather than discrete form. Motion capture was translated into vector and distance-based fields, allowing velocity, acceleration, and directional change to drive fluid particle behavior and time-based trails. Through iteration, particles proved especially effective for making continuity legible—preserving temporal coherence where geometry tends to fragment it.

Reflections

While Tempo made force and continuity visible, the interaction remained indirect: the system responded to recorded motion, but the performer received no real-time feedback while moving. This clarified a key gap—visualizing motion is not the same as interacting with it. Without a feedback loop, motion could be traced and analyzed, but not explored as an ongoing dialogue between body and system.

Emerging question

What happens when motion is not only traced, but answered? How might a system respond as movement unfolds, so the body can adapt, explore, and co-create in real time?

( This shift directly informed Echoes of Motion, where continuity and gesture are experienced through responsive feedback between human movement and computational behavior. )

Future Opportunities

Pathways for further exploration:

From Individual Gesture to Collective Expression
Extend force-based motion mapping into a richer vocabulary to allow multiple bodies to shape overlapping and collective motion fields.

Learning-Based Gesture Interpretation
Explore adaptive systems that learn from repeated movement patterns, enabling the system to respond to personal or evolving motion styles rather than predefined gestures.

Architectural and Multimodal Extensions
Translate force flow into multimodal outputs (sound, light, spatial feedback).