Smart_scatter
Houdini Digital Asset (smart_scatter) that builds a figure from a text or a picture: it places an exact number of drone points along the strokes or outlines of letters set in a font, or along the color boundaries of an image. Corners and joints of the shape always get a point, so the figure stays readable with few drones.
It is a starting point for creating figures. For a text, the points carry the glyph index in glyph_id, which Simple_animation can use to animate letter by letter. The second output shows the source — the set text or the picture — for comparison.
HDA: houdinitools/hda/smart_scatter.hda
Parameters
| Parameter | Type | Default | Description |
|---|---|---|---|
Type type | Menu | Font | What the drone points are placed along. Font: the letters of Text. Silhouette: the color boundaries of a picture. |
Text
Hidden when type != 0
| Parameter | Type | Default | Description |
|---|---|---|---|
Option option | Menu | SingleLine | SingleLine: one point line along the pen stroke of each letter, in drawing order. Countur: points along the outline of each letter. |
Font path font_file | File | — | Font file (.ttf or .otf) the text is set in. |
Icon
| Parameter | Type | Default | Description |
|---|---|---|---|
Image path image_path | File | — | Picture whose color boundaries the points follow. PNG, JPEG, GIF, BMP, WebP and TIFF are read. Hidden when type != 1 |
Settings
| Parameter | Type | Default | Description |
|---|---|---|---|
Total pts count total_pts_count | Integer | 150 | Exact number of points, that is of drones. With too few points for the corners and joints of the shape the node fails instead of dropping them. |
Serif restore serif_restore | Toggle | On | Trace a serif font with its serifs. A kept serif is drawn, so a word needs more strokes and more points; a font without serifs looks the same either way. Hidden when option != 0 or type != 0 |
Text text | String | — | Text to place the points along. Hidden when type != 0 |
Threshold img_threshold | Integer | 20 | How different two colors have to be, as a distance in RGB, for their areas to count as separate shapes. Raise it for a soft or JPEG picture whose edges blend. Hidden when type != 1 |
Min region min_region_area | Float | 0.1 | Smallest shape traced, in thousandths of the picture area (0.1 is 0.01 %). Smaller shapes merge into their neighbour, which keeps JPEG noise out of the formation. Hidden when type != 1 |