Photo Colorization

Transform black and white photographs into vibrant, realistic color images using each::sense. This skill uses AI to intelligently analyze image content and apply historically accurate, contextually appropriate colors.

Features

Automatic Colorization: AI detects subjects and applies natural colors
Historical Accuracy: Context-aware coloring for period-appropriate results
Portrait Enhancement: Realistic skin tones for faces and portraits
Landscape Restoration: Natural colors for outdoor and nature scenes
Vintage Photo Revival: Bring old family photos back to life
Batch Processing: Colorize multiple photos in sequence
Style Preservation: Maintains original photo quality and details

Quick Start

bash

curl -X POST https://sense.eachlabs.run/chat \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "message": "Colorize this black and white photo with natural, realistic colors",
    "image_urls": ["https://example.com/bw-photo.jpg"],
    "mode": "max"
  }'

Common Use Cases

Use Case	Description	Best For
Family Archives	Colorize old family portraits and snapshots	Personal restoration
Historical Documentation	Period-accurate colorization for archives	Museums, historians
Portrait Enhancement	Realistic skin tones and facial features	Professional restoration
Landscape Revival	Natural outdoor and scenic colorization	Nature photography
Vintage Fashion	Accurate clothing and fabric colors	Fashion archives
Military/War Photos	Historical military photograph restoration	Memorial projects
Architecture	Building and interior colorization	Real estate, heritage

Use Case Examples

1. Automatic Colorization

Basic colorization with AI-detected colors.

bash

curl -X POST https://sense.eachlabs.run/chat \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "message": "Colorize this black and white photograph. Apply natural, realistic colors based on the image content. Maintain the original quality and details.",
    "image_urls": ["https://example.com/old-photo.jpg"],
    "mode": "max"
  }'

2. Historical Photo Colorization

Colorize with period-appropriate historical accuracy.

bash

curl -X POST https://sense.eachlabs.run/chat \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "message": "Colorize this historical photograph from the 1940s. Use period-accurate colors typical of that era - muted tones, authentic clothing colors, and historically appropriate details. This appears to be a street scene.",
    "image_urls": ["https://example.com/1940s-street.jpg"],
    "mode": "max"
  }'

3. Family Portrait Colorization

Bring old family photos to life with warm, natural tones.

bash

curl -X POST https://sense.eachlabs.run/chat \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "message": "Colorize this old family portrait photo. Apply natural skin tones, realistic hair colors, and period-appropriate clothing colors. The photo appears to be from the 1950s-1960s era. Make it look warm and inviting.",
    "image_urls": ["https://example.com/family-portrait-bw.jpg"],
    "mode": "max"
  }'

4. Landscape Colorization

Natural outdoor and scenic colorization.

bash

curl -X POST https://sense.eachlabs.run/chat \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "message": "Colorize this black and white landscape photograph. Apply natural colors - blue sky, green foliage, brown earth tones. Make it look like a vibrant summer day while preserving the original composition and atmosphere.",
    "image_urls": ["https://example.com/landscape-bw.jpg"],
    "mode": "max"
  }'

5. Portrait with Specific Skin Tones

Control skin tone colorization for accurate results.

bash

curl -X POST https://sense.eachlabs.run/chat \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "message": "Colorize this portrait photograph. The subject has a medium olive skin tone and dark brown hair. Apply natural, warm skin colors with subtle pink undertones. Eyes appear to be brown. Clothing looks like a dark formal suit.",
    "image_urls": ["https://example.com/portrait-bw.jpg"],
    "mode": "max"
  }'

6. Vintage Photo Colorization

Restore old photographs with era-specific styling.

bash

curl -X POST https://sense.eachlabs.run/chat \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "message": "Colorize this vintage photograph from the 1920s. Apply colors that match the Art Deco era - rich jewel tones for clothing, sepia-influenced skin tones, and period-appropriate interior colors. Maintain the vintage aesthetic while adding realistic color.",
    "image_urls": ["https://example.com/1920s-vintage.jpg"],
    "mode": "max"
  }'

7. War/Military Photo Colorization

Historical military photograph restoration with accuracy.

bash

curl -X POST https://sense.eachlabs.run/chat \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "message": "Colorize this World War II military photograph. Apply historically accurate colors - olive drab for US Army uniforms, appropriate skin tones for the soldiers, realistic equipment colors. Maintain the somber, documentary feel of the original while adding authentic military colors.",
    "image_urls": ["https://example.com/wwii-soldiers.jpg"],
    "mode": "max"
  }'

8. Architecture Colorization

Building and interior photograph colorization.

bash

curl -X POST https://sense.eachlabs.run/chat \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "message": "Colorize this black and white photograph of a Victorian-era building. Apply appropriate architectural colors - red or brown brick, white trim, dark roof tiles, green copper patina if visible. Include natural sky colors and surrounding landscape elements.",
    "image_urls": ["https://example.com/victorian-building.jpg"],
    "mode": "max"
  }'

9. Fashion/Clothing Colorization

Accurate fabric and fashion colorization.

bash

curl -X POST https://sense.eachlabs.run/chat \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "message": "Colorize this 1960s fashion photograph. Apply era-appropriate colors - bold mod colors like orange, turquoise, and pink were popular. The model appears to be wearing a structured dress. Apply realistic fabric textures and vibrant 60s palette.",
    "image_urls": ["https://example.com/1960s-fashion.jpg"],
    "mode": "max"
  }'

10. Batch Colorize Multiple Photos

Process multiple photos with consistent styling using session continuity.

bash

# First photo in batch
curl -X POST https://sense.eachlabs.run/chat \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "message": "Colorize this black and white family photo from the 1950s. Apply warm, natural colors with period-appropriate tones. This is photo 1 of a series from the same family album.",
    "image_urls": ["https://example.com/family-album-001.jpg"],
    "session_id": "family-album-colorization",
    "mode": "max"
  }'

# Second photo (same session for consistency)
curl -X POST https://sense.eachlabs.run/chat \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "message": "Colorize this second photo from the same family album. Maintain consistent skin tones and color style as the previous colorization.",
    "image_urls": ["https://example.com/family-album-002.jpg"],
    "session_id": "family-album-colorization",
    "mode": "max"
  }'

# Third photo (continuing series)
curl -X POST https://sense.eachlabs.run/chat \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "message": "Colorize the third photo. Same family, same era. Keep the color palette consistent with previous images.",
    "image_urls": ["https://example.com/family-album-003.jpg"],
    "session_id": "family-album-colorization",
    "mode": "max"
  }'

Best Practices

Image Quality

Resolution: Higher resolution photos yield better colorization results
Contrast: Good contrast in the original helps AI identify details
Clarity: Sharp, clear photos colorize better than blurry ones
Condition: Digital restoration before colorization can improve results

Prompt Tips

Provide context: Mention the era, location, or subject matter
Specify skin tones: If known, describe the subject's complexion
Note clothing colors: If you know original colors, include them
Describe the scene: Outdoor/indoor, season, time of day

Historical Accuracy

Research the time period for authentic color palettes
Consider regional variations in fashion and architecture
Military uniforms and equipment have specific documented colors
Vintage car and vehicle colors are often well-documented

Mode Selection

Mode	Best For	Speed	Quality
`max`	Final restorations, archival work, gifts	Slower	Highest
`eco`	Quick previews, batch testing, drafts	Faster	Good

Recommendation: Use

eco

mode for initial tests, then

max

for final output.

Multi-Turn Refinement

Use

session_id

to iterate and refine colorization:

bash

# Initial colorization
curl -X POST https://sense.eachlabs.run/chat \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "message": "Colorize this old portrait photograph with natural colors",
    "image_urls": ["https://example.com/portrait.jpg"],
    "session_id": "portrait-restoration"
  }'

# Refine the result
curl -X POST https://sense.eachlabs.run/chat \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "message": "The skin tones look a bit too warm. Can you redo the colorization with cooler, more natural skin tones?",
    "session_id": "portrait-restoration"
  }'

# Further adjustment
curl -X POST https://sense.eachlabs.run/chat \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "message": "The dress should be blue, not green. Please adjust the clothing color.",
    "session_id": "portrait-restoration"
  }'

Error Handling

Error	Cause	Solution
`Failed to create prediction: HTTP 422`	Insufficient balance	Top up at eachlabs.ai
`Invalid image URL`	URL not accessible	Ensure image URL is publicly accessible
Timeout	Large image or complex scene	Set client timeout to minimum 10 minutes
Poor colorization	Low quality source	Use higher resolution, clearer source image

Related Skills

```
each-sense
```
- Core API documentation
```
image-restoration
```
- General image restoration
```
image-enhancement
```
- Enhance and upscale images
```
face-restoration
```
- Specialized face enhancement

photo-colorization

NPX Install

Tags

SKILL.md Content

Photo Colorization

Features

Quick Start

Common Use Cases

Use Case Examples

1. Automatic Colorization

2. Historical Photo Colorization

3. Family Portrait Colorization

4. Landscape Colorization

5. Portrait with Specific Skin Tones

6. Vintage Photo Colorization

7. War/Military Photo Colorization

8. Architecture Colorization

9. Fashion/Clothing Colorization

10. Batch Colorize Multiple Photos

Best Practices

Image Quality

Prompt Tips

Historical Accuracy

Mode Selection

Multi-Turn Refinement

Error Handling

Related Skills