A place to remember, made one small gesture at a time.
I created OFRENDA as an independent 3D experience inspired by Mexico’s Día de Muertos. Visitors build a personal altar with candles, familiar objects, a photograph and marigolds, then save an image to keep or share.
A place built through small gestures
I began with a brainstorm and an experience map: what the visitor would do, what each object meant, and how the camera and words would guide them. Research into Día de Muertos and references from real ofrendas, architecture and lighting informed the setting. The final experience focuses on making a personal altar.
Making the things we leave
My Blender work included food, candles and decorative objects: pan de muerto, a cazuela with rice, sugar skulls and cempasúchil. Their details mattered at close range, from the pot’s painted decoration to the skulls’ icing.
Getting them into the browser meant checking materials, dimensions and contact with the altar. Procedural materials needed baking, and the skull decoration needed exportable geometry. I reviewed the results in browser studies and adjusted them there. The water pitcher uses a separately credited BlenderKit asset.
Offering models together in an early browser development study.
A material close-up used to review the cazuela and rice in the browser.
From Blender to the browser
The cloth, papel picado and falling crates carry authored Blender animation into the experience. I worked through the export and playback so the fabric could settle and the paper could unfold when the visitor requested each step.
React Three Fiber and Three.js handle the interactive scene, with GSAP coordinating transitions. Browser iteration also shaped object placement: grabbing something away from its center and dragging a flower over an edge needed to feel continuous. I refined the framing and mobile scroll distance through repeated visual review.
An upper-tier cloth test, before the final scene lighting.
Development recording of the upper-tier crates settling into place.
Giving the finished scene time
The card-based interface lets visitors choose offerings, add an optional photograph and try different marigold arrangements. Warm lighting and measured camera moves give those choices room to register. English and Spanish copy accompany the same journey.
Browser-local drafts keep the arrangement available to return to. At the end, visitors can save a portrait or landscape image, or share it where the browser allows. The image carries the result of the small choices they made along the way.