PetShimmy

Pet photo guide

How to choose a pet photo for an AI dance video.

Motion transfer can only interpret what the source image shows. Give it one visible pet, a portrait composition, clear light, and enough body detail to understand the subject before it moves.

Published and reviewed: 31 July 2026.

Five-second check

Can you clearly see one face and one body?

  • One pet only
  • Face, body, and paws visible
  • Portrait—not square or landscape

The strongest starting point

Make the animal easy to read before asking it to move.

These are not aesthetic rules for a perfect portrait. They reduce ambiguity for a model that must identify the pet and reinterpret human choreography around its shape.

Show one pet from face to paws

Use a frame where the face, torso, and legs are visible. A close-up headshot gives the model little information for a full-body dance; a hand, blanket, or furniture covering the body can create the same ambiguity.

Choose even, useful light

The pet should stand apart from the background without one side disappearing into shadow. Natural window light or bright outdoor shade often gives clearer fur and facial boundaries than a dark room or harsh flash.

Prefer a calm, sharp frame

Motion blur, screenshot overlays, filters, and heavy compression hide detail the model cannot recover. The pet can sit or stand; visibility matters more than a dramatic pose.

Crop and file requirements

Portrait shape is a product requirement.

PetShimmy validates the image before payment so an unsupported file does not enter the generation pipeline.

If the original is landscape

Crop a portrait version with the pet near the centre, but keep the full ears, face, torso, legs, and paws inside the frame. Do not stretch the photo or fill missing space with a screenshot border.

The exact accepted files

Upload a JPEG, PNG, or WebP under 15 MB. Width and height must each be at least 480 pixels, height must be greater than width, and extremely large or malformed images are rejected.

Do not add a second pet just for context. The current product is designed to animate one subject from one image. A second face or overlapping body makes that subject choice less clear.

Before pressing upload

A final six-point check.

  • There is one pet in the image.
  • The pet’s face, body, legs, and paws are visible.
  • Nothing important is covered by a hand, blanket, or furniture.
  • The photo is portrait, sharp, and evenly lit.
  • The pet is visually distinct from the background.
  • The file is JPEG, PNG, or WebP and no larger than 15 MB.

Photo ready?

Pick the dance before you pay.

Watch five genuine AI-generated pet outputs, hear their matching soundtracks, and choose the movement that fits.

Compare the dances