Skip to content
The lab/Motion Sketch
Home ↗
Connecting to rendererHow the pharmacist works →

REQUESTED VS. DELIVERED LIVE EXPERIMENT

The pharmacist
and the generator.

The pharmacist turns each request into a scene a renderer can execute. The generator renders the image pair with correspondence labels.

01 Sketch02 Scene03 Pair + labels
01 / INPUT

Sketch the motion

0 arrows

Draw your own arrows, or load a preset below. Presets are starting points you can edit.

A → B 512 × 512
Drag from start to destination.

A sketch is a request, not a guarantee. Compare the motion you asked for with what the scene delivers.

02 / OUTPUT

Meet your scene

AWAITING A SKETCH
Draw arrows on the canvas, or load a preset to get started.
A

The starting view

Frame ACorrespondences start here
B

The destination view

Frame BThe same scene, moved
Hover or touch frame A to follow a point.
Requested motionpx median
Delivered foregroundpx median
Valid label coverageof frame A

Requested vs. delivered motion

A → B · SAME PIXEL SCALE

Input arrows sit on top of a sparse sample of valid foreground flow. Both use their original positions and lengths.

Generate a pair to compare its labels with your sketch.

Dense flowColor = direction · brightness = distance
Validity maskWhite = a visible match in both views
TAKE IT WITH YOU

Two images.
Pixel-level labels.
Your scene recipe.

Download bundle ↓Labels only (.npz)Generated on your local GPU
WHAT YOU ARE SEEING

Ground each request
in a realizable scene.

The arrows describe positions and displacements. A trained inverse model—the pharmacist—maps this set to camera and object parameters. The renderer makes the pair and computes correspondences from the scene’s surfaces.

These are synthetic scenes from the research pipeline. Labels describe the actual rendered motion, including visibility; they do not copy your sketch. Arbitrary drawings may fall outside what the model can realize.