A two-person costume concept that works as a couple photo, not two unrelated outfits. One western silhouette, one feathered angel silhouette, and a shared desert-night scene.
Updated October 20, 2026

Eight original looks built around this year's trending costume language: gothic braids, celestial princess, sci-fi visor hero, pumpkin pet, robot, haunted mascot, liminal horror, and a cowboy-angel couple.
Use a front-facing photo with even light and no sunglasses. One person works best; couples need a multi-reference model.
The prompt protects the face, skin tone, body proportions, and the number of people. Change the costume and scene, not the person.
Generate the still first. When the face and costume look right, send that image to an image-to-video model for a five-second clip.
This is the only look in the set that needs two reference photos. Generate a solo version first if you want to test the lighting before adding the second person.
Open the image generator, upload your photo, and keep the identity-preservation lines intact. The prompt is written for an image-editing model, not a generic text-to-image prompt.
Create a photorealistic editorial portrait of two adults in a coordinated cowboy-and-angel Halloween couple costume. One wears a dark western jacket, silver belt buckle, black hat and dusty boots; the other wears a soft ivory draped outfit with feathered wings and a subtle silver halo. Both stand in a desert motel courtyard at night with warm neon spill, a vintage pickup truck, and a clear starry sky. Keep both faces recognizable and natural. The mood is cinematic and stylish, not costume-shop cheesy. No text, no logos, no watermark.
Each look can run on more than one model. Start with an identity-first model, then compare a scene-first edit when you need a stronger costume or background.
Yes, use a multi-reference model and label each photo. Say which person wears the cowboy outfit and which wears the angel outfit so the model does not swap them.
The prompt says two adults only. If a third person appears, repeat "exactly two people" and remove any group language from the prompt.
Keep the materials realistic and the poses restrained. Ask for "editorial fashion photography" and avoid props like plastic halos or toy guns.
Generate the still first. When the face and costume look right, send that image to an image-to-video model for a five-second clip.
Original AI interpretations. Vidle is not affiliated with film, game, or costume rights holders. Only upload photos of people who have given you permission.