AI Photo Colorization

AI Photo Colorization

Colorize black and white photographs with realistic, natural colors using each::sense — the intelligent AI agent that automatically selects the best model for your request.

Quick Start

Requires an each::labs API key. Get one at eachlabs.ai.

Using curl

curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -d '{
    "messages": [
      {
        "role": "user",
        "content": [
              {"type": "text", "text": "Colorize this black and white photograph with realistic natural colors: accurate skin tones, believable clothing colors appropriate to the era, natural sky and vegetation colors, maintain the original photographic quality and composition"},
              {"type": "image_url", "image_url": {"url": "https://example.com/black-and-white-photo.jpg"}}
            ]
          }
    ],
    "stream": false
  }'

Using Python (OpenAI SDK)

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_EACHLABS_API_KEY",
    base_url="https://eachsense-agent.core.eachlabs.run/v1"
)

response = client.chat.completions.create(
    model="eachsense/beta",
    messages=[{
        "role": "user",
        "content": "Colorize this black and white photograph with realistic natural colors: accurate skin tones, believable clothing colors appropriate to the era, natural sky and vegetation colors, maintain the original photographic quality and composition"
    }],
    # Images are included in the message content array above
)

print(response.choices[0].message.content)

Streaming

Set "stream": true for real-time SSE responses, or "stream": false for complete result in a single response. Streaming is useful for showing progress in UIs; non-streaming is simpler for scripts and automation.

Tips for Photo Colorization

• Always provide the source image via image_urls — colorization requires the grayscale original.
• Specify the era — "1940s wartime," "1960s," "Victorian era" helps the model choose period-appropriate colors.
• Guide specific colors when known — "the car was red," "she wore a blue dress," "the house had green shutters."
• Mention the scene type — portrait, landscape, street scene, or interior to improve color accuracy.
• Request subtle or vibrant — "muted period-accurate tones" vs. "vibrant saturated colors."

Examples

Portrait Colorization

curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -d '{
    "messages": [
      {
        "role": "user",
        "content": [
              {"type": "text", "text": "Colorize this 1940s black and white portrait: natural warm skin tones, the subject is wearing a dark navy suit with a white shirt, brown hair, soft warm studio lighting tones, keep the classic portrait feel while adding realistic color"},
              {"type": "image_url", "image_url": {"url": "https://example.com/vintage-portrait-bw.jpg"}}
            ]
          }
    ],
    "stream": false
  }'

Historical Street Scene

curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -d '{
    "messages": [
      {
        "role": "user",
        "content": [
              {"type": "text", "text": "Colorize this 1920s city street photograph: period-accurate colors for brick buildings in warm red-brown, grey cobblestone streets, vintage automobiles in dark greens and blacks, pedestrians in muted earth-toned clothing, overcast sky, historically authentic color treatment"},
              {"type": "image_url", "image_url": {"url": "https://example.com/1920s-street-bw.jpg"}}
            ]
          }
    ],
    "stream": false
  }'

Landscape Colorization

curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -d '{
    "messages": [
      {
        "role": "user",
        "content": [
              {"type": "text", "text": "Colorize this black and white landscape photograph: lush green forests, blue sky with white clouds, brown earth tones on the mountain face, golden sunlight on the meadow in the foreground, natural vivid colors, photorealistic colorization"},
              {"type": "image_url", "image_url": {"url": "https://example.com/landscape-bw.jpg"}}
            ]
          }
    ],
    "stream": false
  }'

Workflow: Archival Colorization Pipeline

1. Restore the source — Fix any damage before colorization using photo restoration.
2. Research the era — Identify the time period and likely colors for accuracy.
3. Provide color guidance — Specify known colors for clothing, vehicles, buildings, and signage.
4. Colorize the image — Apply AI colorization with era-specific cues.
5. Review and adjust — Check for color bleeding, unnatural tones, and historical accuracy.
6. Enhance the result — Apply HDR enhancement for final polish.

Related Skills

• AI Photo Restoration — Repair damage before colorization
• AI HDR Enhancement — Enhance dynamic range after colorization
• AI Photo Style Transfer — Apply artistic color treatments
• AI Stock Photo — Generate color stock photography

Related Models

• flux-2-max — Accurate, detailed colorization
• seedream-v4-5 — Intelligent color inference
• nano-banana-pro — Quick colorization previews

Documentation

• each::sense Overview
• each::labs API