Motion Control Collection is an ongoing series of AI-generated video experiments. The project explores the relationship between human movement, synthetic characters, and generative filmmaking.

The project begins with real-world movement. I collected footage of trending dances from social media and used the performances as motion references and choreographic foundations. I then developed original characters as individual still images using Midjourney and Kling. This process allowed me to experiment with character design, fashion, visual identity, and photographic style.

Next, I brought the still characters to life with Kling 3.0 Motion Control. The reference footage guided their movements, gestures, timing, and physical rhythm. Kling translated these performances into new characters and visual worlds. The resulting videos retain traces of human choreography while transforming the performer, context, and aesthetic through generative AI.

The collection functions as both a creative study and a technical experiment. It explores how human motion can become a directing tool for synthetic performance. Instead of relying entirely on text prompts to describe movement, I use source footage as a form of embodied instruction. It directs gesture, posture, rhythm, timing, and choreography while AI interprets the performance through a new visual identity.

Across the series, Motion Control Collection explores the space between documentation and fabrication, as well as human performance and machine interpretation. The project asks a central question: What happens when we separate movement from its original body and give it an entirely new visual identity?