Hugging Face Image - AI Image Processing with Stable Diffusion & Vision Models

Generate, transform, analyze, and understand images with state-of-the-art AI models. Stable Diffusion XL for generation, BLIP for captioning, ViT for classification, DETR for object detection, and more. No GPU required - access the best vision models through API. BYOK with your Hugging Face API token.

Features

• Text-to-Image - Generate images with Stable Diffusion XL
• Image-to-Image - Transform images with prompts
• Inpainting - Edit specific parts of images
• Image Captioning - Generate descriptions with BLIP
• Image Classification - Identify objects with Vision Transformer
• Object Detection - Locate objects with DETR
• Image Segmentation - Pixel-level scene analysis
• Zero-Shot Classification - Classify images without training
• Depth Estimation - Generate depth maps
• Image Upscaling - 4x super-resolution
• Demo Mode - Test with sample data before going live

Who Should Use This Actor?

Marketing Teams

Generate product images. Create social media visuals. Caption images for SEO. A/B test creative variants.

E-commerce Businesses

Generate product photos. Remove backgrounds. Upscale images. Auto-caption for accessibility.

Content Creators

Generate blog illustrations. Transform stock photos. Create custom visuals without designers.

Developers

Add AI image features to apps. Build image analysis pipelines. Integrate vision AI without infrastructure.

Social Media Managers

Generate post images. Analyze image content. Auto-tag visual content.

Research Teams

Analyze image datasets. Generate synthetic data. Extract visual features.

Quick Start

Demo Mode (Free Test)

{
  "task": "text_to_image",
  "prompt": "Professional product photo of sleek wireless earbuds on white background",
  "demoMode": true
}

Text-to-Image (Stable Diffusion)

{
  "task": "text_to_image",
  "apiToken": "hf_your_token_here",
  "model": "stabilityai/stable-diffusion-xl-base-1.0",
  "prompt": "Modern minimalist office interior, natural lighting, 4k, professional photography",
  "negativePrompt": "blurry, low quality, distorted",
  "width": 1024,
  "height": 1024,
  "guidanceScale": 7.5,
  "numInferenceSteps": 50,
  "demoMode": false
}

Image-to-Image Transformation

{
  "task": "image_to_image",
  "apiToken": "hf_your_token_here",
  "imageUrl": "https://example.com/photo.jpg",
  "prompt": "Transform into watercolor painting style",
  "strength": 0.8,
  "guidanceScale": 7.5,
  "demoMode": false
}

Inpainting (Edit Parts of Images)

{
  "task": "inpainting",
  "apiToken": "hf_your_token_here",
  "imageUrl": "https://example.com/original.jpg",
  "maskUrl": "https://example.com/mask.png",
  "prompt": "A red sports car",
  "guidanceScale": 7.5,
  "demoMode": false
}

Image Captioning

{
  "task": "image_to_text",
  "apiToken": "hf_your_token_here",
  "imageUrl": "https://example.com/product-photo.jpg",
  "demoMode": false
}

Image Classification

{
  "task": "image_classification",
  "apiToken": "hf_your_token_here",
  "imageUrl": "https://example.com/photo.jpg",
  "demoMode": false
}

Object Detection

{
  "task": "object_detection",
  "apiToken": "hf_your_token_here",
  "imageUrl": "https://example.com/street-scene.jpg",
  "demoMode": false
}

Zero-Shot Image Classification

{
  "task": "zero_shot_image_classification",
  "apiToken": "hf_your_token_here",
  "imageUrl": "https://example.com/photo.jpg",
  "candidateLabels": "product photo,lifestyle image,infographic,screenshot",
  "demoMode": false
}

Depth Estimation

{
  "task": "depth_estimation",
  "apiToken": "hf_your_token_here",
  "imageUrl": "https://example.com/scene.jpg",
  "demoMode": false
}

Image Upscaling (4x)

{
  "task": "image_upscaling",
  "apiToken": "hf_your_token_here",
  "imageUrl": "https://example.com/low-res.jpg",
  "prompt": "high quality, detailed, sharp",
  "demoMode": false
}

Input Parameters

Parameter	Type	Default	Description
task	string	required	Task to perform (see task list)
apiToken	string	-	Your Hugging Face API token
model	string	task default	Specific model to use
prompt	string	-	Text prompt for generation
negativePrompt	string	-	What to avoid in generation
imageUrl	string	-	Input image URL
maskUrl	string	-	Mask image URL (for inpainting)
candidateLabels	string	-	Comma-separated labels (zero-shot)
width	number	1024	Output image width
height	number	1024	Output image height
guidanceScale	number	7.5	CFG scale (prompt adherence)
numInferenceSteps	number	50	Diffusion steps
strength	number	0.8	Transformation strength (0-1)
seed	number	-	Random seed for reproducibility
scheduler	string	"default"	Diffusion scheduler
waitForModel	boolean	true	Wait for model to load
webhookUrl	string	-	Webhook URL for results
demoMode	boolean	true	Return sample data

Available Tasks

Task	Description	Default Model
text_to_image	Generate images from text	Stable Diffusion XL
image_to_image	Transform images	SDXL Refiner
inpainting	Edit image regions	Stable Diffusion XL
image_to_text	Generate captions	BLIP-large
image_classification	Classify images	ViT-base-patch16
object_detection	Detect objects	DETR-ResNet-50
image_segmentation	Segment scenes	DETR-panoptic
zero_shot_image_classification	Classify with custom labels	CLIP-ViT-large
depth_estimation	Generate depth maps	DPT-large
image_upscaling	4x super-resolution	SD-x4-upscaler

Output Format

Text-to-Image

{
  "success": true,
  "model": "stabilityai/stable-diffusion-xl-base-1.0",
  "imageBase64": "iVBORw0KGgoAAAANSUhEUgAAA...",
  "mimeType": "image/png",
  "prompt": "Modern minimalist office interior..."
}

Image Captioning

{
  "success": true,
  "model": "Salesforce/blip-image-captioning-large",
  "caption": "A sleek pair of wireless earbuds displayed on a white background with soft shadows",
  "imageUrl": "https://example.com/product-photo.jpg"
}

Image Classification

{
  "success": true,
  "model": "google/vit-base-patch16-224",
  "classifications": [
    {"label": "wireless headphones", "score": 0.92},
    {"label": "earbuds", "score": 0.06},
    {"label": "hearing aid", "score": 0.01}
  ],
  "imageUrl": "https://example.com/photo.jpg"
}

Object Detection

{
  "success": true,
  "model": "facebook/detr-resnet-50",
  "objects": [
    {"label": "person", "score": 0.98, "box": {"xmin": 100, "ymin": 50, "xmax": 300, "ymax": 400}},
    {"label": "car", "score": 0.95, "box": {"xmin": 400, "ymin": 200, "xmax": 600, "ymax": 350}},
    {"label": "dog", "score": 0.87, "box": {"xmin": 150, "ymin": 300, "xmax": 250, "ymax": 420}}
  ],
  "imageUrl": "https://example.com/street-scene.jpg"
}

Image Segmentation

{
  "success": true,
  "model": "facebook/detr-resnet-50-panoptic",
  "segments": [
    {"label": "person", "score": 0.95, "mask": "base64_encoded_mask..."},
    {"label": "sky", "score": 0.98, "mask": "base64_encoded_mask..."},
    {"label": "grass", "score": 0.92, "mask": "base64_encoded_mask..."}
  ]
}

Depth Estimation

{
  "success": true,
  "model": "Intel/dpt-large",
  "depthMapBase64": "iVBORw0KGgoAAAANSUhEUgAAA...",
  "mimeType": "image/png"
}

Pricing (Pay-Per-Event)

Event	Description	Price
image_processed	Per image task completed	$0.01

Example costs:

• 50 image generations: 50 × $0.01 = $0.50
• 100 image classifications: 100 × $0.01 = $1.00
• 200 captions generated: 200 × $0.01 = $2.00
• Demo mode: $0.00

Note: Hugging Face Pro may be required for some models

Cost Comparison

Tool	Per Image	This Actor
Midjourney	~$0.10	~$0.01
DALL-E 3	~$0.04	~$0.01
Leonardo.ai	~$0.05	~$0.01

Common Scenarios

Scenario 1: Product Image Generation

{
  "task": "text_to_image",
  "apiToken": "hf_your_token",
  "prompt": "Professional product photography of premium leather wallet, studio lighting, white background, 4k quality",
  "negativePrompt": "blurry, low quality, watermark, text",
  "width": 1024,
  "height": 1024,
  "guidanceScale": 8.0,
  "seed": 42,
  "demoMode": false
}

Scenario 2: Auto-Generate Alt Text

{
  "task": "image_to_text",
  "apiToken": "hf_your_token",
  "imageUrl": "https://example.com/blog-header.jpg",
  "webhookUrl": "https://hooks.zapier.com/...",
  "demoMode": false
}

Scenario 3: Content Moderation

{
  "task": "zero_shot_image_classification",
  "apiToken": "hf_your_token",
  "imageUrl": "https://example.com/user-upload.jpg",
  "candidateLabels": "safe content,adult content,violence,spam",
  "demoMode": false
}

Scenario 4: Style Transfer

{
  "task": "image_to_image",
  "apiToken": "hf_your_token",
  "imageUrl": "https://example.com/photo.jpg",
  "prompt": "Oil painting in the style of Van Gogh, impressionist, vibrant colors",
  "strength": 0.75,
  "demoMode": false
}

Webhook & Automation Integration

Zapier / Make.com / n8n

1. Create a webhook trigger
2. Copy the URL to webhookUrl
3. Process image results in your workflow

Popular automations:

• Generated images -> Cloud storage upload
• Captions -> CMS alt text updates
• Object detection -> Inventory tagging
• Classification -> Content moderation queue

Hugging Face AI Suite

Actor	Best For
Hugging Face Master	All-in-one (text + image + audio)
Hugging Face Text	Text processing
Hugging Face Image	Image processing (lightweight)
Hugging Face Audio	Audio processing
Hugging Face Hub	Model discovery

FAQ

Q: What image formats are supported?

A: JPEG, PNG, WebP. Output is always PNG for quality.

Q: What's the max image size?

A: Input images up to 10MB. Generation up to 1024x1024 (SDXL).

Q: Can I generate multiple images?

A: Run multiple times with different seeds. Use seed parameter for reproducibility.

Q: How do I improve generation quality?

A: Increase numInferenceSteps (50-100), tune guidanceScale (7-12), use detailed prompts.

Q: What's the difference from DALL-E/Midjourney?

A: Similar quality, pay-per-use pricing, more model choices, API-first.

Common Problems & Solutions

"Model is loading"

• SDXL is large, needs warm-up
• Set waitForModel: true (default)
• Consider smaller models for testing

"Image too large"

• Resize input images before sending
• Use max 1024x1024 for generation

"Low quality output"

• Increase numInferenceSteps to 50+
• Use negative prompts to avoid artifacts
• Try different guidanceScale values

"Demo data showing"

• Set demoMode: false
• Provide your Hugging Face API token

---

Built by John Rippy | Actor Arsenal