brain_deer.surface.figures¶

Publication output for Surface Studio: multi-view figures, video, glTF, presets.

export_multiview_figure() renders the scene from several canonical views and lays them out in a labelled grid with a shared colour bar (PNG/SVG, high-DPI).
export_rotation_video() spins the camera and writes a video (optional imageio) or a numbered PNG frame sequence as a fallback.
export_scene_gltf() writes the current scene to glTF for Blender.
Scene presets (scene_to_preset() / apply_preset()) save & restore the presentation settings (colormap, vmin/vmax, threshold, views, background) as JSON, without the heavy geometry.

The pure compositor (compose_grid()) and preset (de)serialisation are numpy/JSON only and unit-tested headlessly; the rendering entry points need a live viewer.

Functions¶

`compose_grid`() → numpy.ndarray)	Tile RGB images (H,W,3 uint8) into a padded grid; cells are size-normalised.
`export_multiview_figure`(, dpi, ncols, show_labels)	Render `scene` from `views` and save a labelled grid figure (PNG/SVG).
`export_rotation_video`(, fps, axis)	Spin the camera 360° and save a video (imageio) or PNG frame sequence.
`export_scene_gltf`(→ str)	Export the current surface scene to a glTF file for Blender/other engines.
`scene_to_preset`(→ dict)	Serialise presentation settings (not geometry) to a JSON-able dict.
`apply_preset`(→ brain_deer.surface.scene_model.SceneModel)	Apply a preset's presentation settings onto an existing scene in place.

Module Contents¶

brain_deer.surface.figures.compose_grid(images: Sequence[numpy.ndarray], *, ncols: int | None = None, pad: int = 12, bg: tuple[int, int, int] = (255, 255, 255)) → numpy.ndarray¶

Tile RGB images (H,W,3 uint8) into a padded grid; cells are size-normalised.

All cells use the max height/width across inputs so ragged sizes still align.

brain_deer.surface.figures.export_multiview_figure(viewer, scene: brain_deer.surface.scene_model.SceneModel, path: str, *, views: Sequence[str] | None = None, size: tuple[int, int] = (700, 700), dpi: int = 200, ncols: int | None = None, show_labels: bool = True) → str¶: Render scene from views and save a labelled grid figure (PNG/SVG).

brain_deer.surface.figures.export_rotation_video(viewer, path: str, *, n_frames: int = 36, size: tuple[int, int] = (700, 700), fps: int = 20, axis: str = 'azimuth') → str¶: Spin the camera 360° and save a video (imageio) or PNG frame sequence.

brain_deer.surface.figures.export_scene_gltf(viewer, path: str) → str¶: Export the current surface scene to a glTF file for Blender/other engines.

brain_deer.surface.figures.scene_to_preset(scene: brain_deer.surface.scene_model.SceneModel) → dict¶: Serialise presentation settings (not geometry) to a JSON-able dict.

brain_deer.surface.figures.apply_preset(scene: brain_deer.surface.scene_model.SceneModel, preset: dict) → brain_deer.surface.scene_model.SceneModel¶: Apply a preset’s presentation settings onto an existing scene in place.