Image to Video API
Animate Any Still Image

How Image-to-Video Works

Image-to-video (also called image animation) takes a still image as the first frame and generates a video that extends naturally from it. The AI model analyzes the composition, subjects, lighting, and depth of your image, then synthesizes realistic motion that is consistent with the original frame.

With US Video API, you pass your image using the first_frame parameter — either as a base64-encoded string or a publicly accessible URL. You also provide a prompt that describes the desired motion: "camera slowly zooms in," "the subject turns to face the camera," or "leaves blow gently in the wind."

Seedance 2.0 is particularly strong at image-to-video. It preserves fine details from the source image — textures, brand colors, text on packaging — while adding natural, physically plausible motion. This makes it ideal for commercial applications where visual accuracy matters.

The first_frame Parameter

The key parameter for image-to-video generation is first_frame. This tells the model to use your image as the starting point of the video rather than generating from scratch.

import requests, base64

# Option A: Image URL
job = requests.post(
    "https://usvideoapi.com/v1/videos",
    headers={"Authorization": f"Bearer {API_KEY}"},
    json={
        "prompt": "Slow zoom in, product rotates slightly, soft studio lighting",
        "first_frame": "https://example.com/product-photo.jpg",
        "resolution": "1080p",
        "duration": 5,
    }
).json()

# Option B: Base64-encoded image
with open("product.jpg", "rb") as f:
    img_b64 = base64.b64encode(f.read()).decode()

job = requests.post(
    "https://usvideoapi.com/v1/videos",
    headers={"Authorization": f"Bearer {API_KEY}"},
    json={
        "prompt": "Camera orbits around the product, dramatic lighting",
        "first_frame": f"data:image/jpeg;base64,{img_b64}",
        "resolution": "1080p",
        "duration": 5,
    }
).json()

Use Cases for Image Animation

Best Practices for Source Images

The quality of your source image directly impacts the quality of the generated video. Follow these guidelines for optimal results:

• High resolution input — Use images at least 1024x1024 pixels. The model downsamples internally but benefits from sharp source material.
• Clean composition — Simple, well-lit images with clear subjects produce the best motion. Cluttered backgrounds can confuse the motion model.
• JPEG or PNG — Both formats are supported. PNG preserves transparency but results in larger payloads. JPEG is recommended for photos.
• Aspect ratio matters — 16:9 for landscape video, 9:16 for vertical/shorts, 1:1 for social. The output video matches the input image aspect ratio.
• Pair with a motion prompt — Always include a prompt describing the desired motion. Without it, the model applies default subtle motion which may not match your intent.

Pricing

Image-to-video uses the same pricing as text-to-video — billed per second of output:

• 480p — $0.10/second
• 720p — $0.25/second
• 1080p — $0.50/second

The first_frame image upload is free — you only pay for the video output. No additional charges for image processing or hosting.