Graphics & computation · 13 studies
Studies
Smaller projects built around a specific idea. Explore the algorithm, change its inputs, and see what happens.

Image quilting
Expanding a texture into a larger surface with overlapping, near-full copies of the source. Minimum-error seams hide the joins between them.

Texture transfer
Reconstructing the shape and tones of one image with the material of another. Every output pixel comes from the source texture.

Electron Orbitals
A Python study of electron probability density, turning radial and angular expressions into high-resolution images.

Weighted stippling
Thousands of dots settle into an image. Darker regions attract more ink; weighted Voronoi cells keep the distribution even.

Painterly rendering
Repaint a photograph with curved strokes, starting with broad shapes and returning for smaller details.

Seam carving
Make an image narrower by removing low-energy paths through it, with painted masks to protect the parts that matter.

PatchMatch completion
Paint out an object and let matching patches from elsewhere in the image fill the gap.

Moving least squares
Pin an image in place, pull one point, and watch the surrounding shape bend through a field of local transformations.

Space-colonization trees
Grow a branching tree toward a cloud of attraction points. Its shape emerges from competition for space.

Stable Fluids
Stir colored dye through a two-dimensional fluid. Diffusion, advection, and pressure projection evolve the field.

Hybrid images
Combine the fine detail of one image with the broad structure of another. What you see changes with viewing distance.

HDR tone mapping
Compress a high-dynamic-range photograph for a regular display while preserving detail in its bright and dark regions.

Image Analogies
Show an algorithm an image and its painted counterpart. It uses that relationship to synthesize a treatment for a different image.
From a paper to a working experiment.
These studies explore image processing, rendering, and simulation through working implementations. Change the inputs, inspect the results, and follow the original research. The interactive algorithms share one source repository; Electron Orbitals preserves an earlier Python project and its render archive. Each study explains what was implemented and where the approach has limits.