Image Pipeline Playground
export type TemplateCategory =
| 'edges'
| 'enhance'
| 'segment'
| 'structure'
| 'augment'
| 'color'
| 'script'
export type WorkflowTemplate = {
id: string
name: string
description: string
/** Public URL path under /examples/ */
path: string
category: TemplateCategory
/** Short pipeline step labels shown in the gallery tile. */
steps: string[]
}
export const TEMPLATE_CATEGORY_META: Record<
TemplateCategory,
{ label: string; accent: string }
> = {
edges: { label: 'Edges', accent: '#e67e22' },
enhance: { label: 'Enhance', accent: '#5dade2' },
segment: { label: 'Segment', accent: '#48c9b0' },
structure: { label: 'Structure', accent: '#af7ac5' },
augment: { label: 'Augment', accent: '#e74c3c' },
color: { label: 'Color', accent: '#f5b041' },
script: { label: 'Script', accent: '#58d68d' },
}
export const WORKFLOW_TEMPLATES: WorkflowTemplate[] = [
{
id: 'blur_canny',
name: 'Blur & Canny',
description: 'Load anna-portrait, blur, then extract Canny edges.',
path: '/examples/blur_canny.json',
category: 'edges',
steps: ['Load', 'Blur', 'Canny'],
},
{
id: 'hist_equalize',
name: 'Hist Equalize',
description: 'Grayscale + histogram equalize for OCR / documents.',
path: '/examples/hist_equalize.json',
category: 'enhance',
steps: ['Load', 'Gray', 'Equalize'],
},
{
id: 'morphology_cleanup',
name: 'Morphology Cleanup',
description: 'Otsu threshold then morphological open to clean noise.',
path: '/examples/morphology_cleanup.json',
category: 'segment',
steps: ['Load', 'Gray', 'Otsu', 'Open'],
},
{
id: 'contour_detect',
name: 'Contour Detect',
description: 'Threshold then find contours for structure analysis.',
path: '/examples/contour_detect.json',
category: 'structure',
steps: ['Load', 'Thresh', 'Contours'],
},
{
id: 'albumentations_augment',
name: 'Albu Augment',
description: 'Annotations + color jitter + flip with synced bboxes/keypoints.',
path: '/examples/albumentations_augment.json',
category: 'augment',
steps: ['Load', 'Annotate', 'Jitter', 'Flip'],
},
{
id: 'albumentations_weather',
name: 'Albu Weather',
description: 'Synthetic fog then rain for weather degradation.',
path: '/examples/albumentations_weather.json',
category: 'augment',
steps: ['Load', 'Fog', 'Rain'],
},
{
id: 'albumentations_noise_blur',
name: 'Albu Noise & Blur',
description: 'Gaussian noise then motion blur — sensor / shake look.',
path: '/examples/albumentations_noise_blur.json',
category: 'augment',
steps: ['Load', 'Noise', 'Motion'],
},
{
id: 'albumentations_geometry',
name: 'Albu Geometry',
description: 'Rotate + affine with annotations kept in sync.',
path: '/examples/albumentations_geometry.json',
category: 'augment',
steps: ['Load', 'Annotate', 'Rotate', 'Affine'],
},
{
id: 'hsv_color_mask',
name: 'HSV Color Mask',
description: 'HSV in-range mask then apply to the original image.',
path: '/examples/hsv_color_mask.json',
category: 'segment',
steps: ['Load', 'HSV', 'In Range', 'Mask'],
},
{
id: 'clahe_sharpen',
name: 'CLAHE & Sharpen',
description: 'Local contrast (CLAHE) followed by sharpening.',
path: '/examples/clahe_sharpen.json',
category: 'enhance',
steps: ['Load', 'CLAHE', 'Sharpen'],
},
{
id: 'kmeans_palette',
name: 'K-Means Palette',
description: 'Quantize colors with k-means for a compact palette.',
path: '/examples/kmeans_palette.json',
category: 'color',
steps: ['Load', 'K-Means'],
},
{
id: 'custom_python_sepia',
name: 'Custom Python Sepia',
description:
'Load anna-portrait, then apply a sepia look via inline Custom Python (requires trust before Run).',
path: '/examples/custom_python_sepia.json',
category: 'script',
steps: ['Load', 'Custom Python'],
},
]