ShadowDraw: From Any Object to Shadow-Drawing Compositional Art
Rundong Luo, Noah Snavely, Wei-Chiu Ma
2025-12-05
Summary
This paper introduces ShadowDraw, a system that automatically creates art by cleverly using shadows. It takes a regular 3D object and figures out the best lighting and viewpoint so that the shadow it casts completes a drawing, making a recognizable image.
What's the problem?
Creating interesting and artistic images with computers is hard, especially when you want something beyond just realistic renderings. Existing methods often lack a way to combine 3D objects with artistic styles like drawing in a meaningful way. The challenge is to make the shadow *part* of the artwork, not just a byproduct of the lighting.
What's the solution?
The researchers developed ShadowDraw, which works in a few key steps. First, it guesses the best position and lighting for the 3D object. Then, it creates a partial line drawing, and crucially, it uses the shadow itself to finish the drawing. They also built a way to automatically check if the resulting image looks good and makes sense as a shadow drawing, refining the process until it does. They tested it with different kinds of 3D models, even making animations and using it with real-world objects.
Why it matters?
This work is important because it offers a new way to create art using computers, blending algorithmic design with artistic expression. It’s not just about making pretty pictures; it expands the possibilities for how we can use technology to help with creative tasks and potentially even tell stories visually. It provides a practical tool for artists and designers to explore a unique style of art.
Abstract
We introduce ShadowDraw, a framework that transforms ordinary 3D objects into shadow-drawing compositional art. Given a 3D object, our system predicts scene parameters, including object pose and lighting, together with a partial line drawing, such that the cast shadow completes the drawing into a recognizable image. To this end, we optimize scene configurations to reveal meaningful shadows, employ shadow strokes to guide line drawing generation, and adopt automatic evaluation to enforce shadow-drawing coherence and visual quality. Experiments show that ShadowDraw produces compelling results across diverse inputs, from real-world scans and curated datasets to generative assets, and naturally extends to multi-object scenes, animations, and physical deployments. Our work provides a practical pipeline for creating shadow-drawing art and broadens the design space of computational visual art, bridging the gap between algorithmic design and artistic storytelling. Check out our project page https://red-fairy.github.io/ShadowDraw/ for more results and an end-to-end real-world demonstration of our pipeline!