Can I generate AI product photos with the model in a specific city or street backdrop?

Quick Answer

Yes. Generate the product on a model against a specific city or street by combining the on-model setup with a reusable scene reference for the location. In Nightjar, switch Image Type to lifestyle, choose a Fashion Model, pick a Photography Style for the city light, set a Background or scene reference for the street, and add a one-line note for the specifics. Stay generic when possible (a Parisian boulevard rather than a named landmark) to avoid likeness and trademark exposure.

How to set the city scene

Nightjar uses a Product Listing Image Workflow, the core Create path that turns a product photo into ecommerce-ready imagery. Five controls do the work:

Image Type. Toggle from listing to lifestyle so the result is the product and model in a real scene, not on a clean backdrop.
Fashion Model. Nightjar's reusable AI person used to appear with a product. Pick from 80+ pre-built Fashion Models or build a custom one. Reuse the same model across an urban campaign so identity stays consistent across SKUs.
Photography Style. Nightjar's reusable visual direction for camera, lighting, color, and mood. Pick a curated outdoor style ("golden-hour street," "overcast city sidewalk," "neon-lit night") or build a custom one from a reference image.
Background. A solid color or a scene image. Drop in a reference photo of the street, or leave Background empty and let the Photography Style and the next field carry the location.
Custom Directions. Plain-English instructions layered on top of the ingredients. This is where specifics go: "Parisian Haussmann boulevard, wet cobblestones, late afternoon," or "downtown Tokyo crosswalk, neon signage, light rain at dusk."

If you already have a photo of the exact street you want, use the Edit tab instead. Add the on-model image as @image1 and the street photo as @image2, then click the Product Placement shortcut. Nightjar fills in the prompt and the model and product inherit the scene's light, color, and shadows.

For catalog scale, save the whole setup as a reusable Create-form bundle so the next SKU lands in the same city without rebuilding the brief.

How specific you can get

The trade-off runs along three bands:

Specificity	Example	Typical result
Generic city character	"European cobblestone street," "Brooklyn brownstone block," "Tokyo backstreet"	Reads as a real city. Reliable.
Named city, generic location	"side street in Paris," "Soho rooftop, New York," "alleyway in Marrakech"	Usually clean. AI models have strong priors for these.
Named landmark	"Eiffel Tower in frame," "Times Square billboards," "Hollywood Sign"	Renders imperfectly, and raises rights questions.

Three things stack against named landmarks. First, generative models render landmark architecture and real signage imperfectly, and wrong real text is more noticeable than slightly-wrong generic architecture. Second, several landmarks carry trademark or controlled-image rights. The Eiffel Tower's nighttime illumination is treated as a protected creative work managed by SETE, the Hollywood Sign is trademarked, and many private buildings restrict commercial photography. Third, visible third-party logos or shop signs in a commercial image can create false-endorsement exposure. If a landmark is load-bearing for the campaign, treat that image as legal review territory, not a default catalog output.

Bystanders fall under the same caution. If a background figure reads as an identifiable real person, the right-of-publicity concerns that apply to AI-generated celebrity models apply to incidental figures too. Keep background characters generic, soft-focus, or remove them.

What makes a street scene look real

The common failure mode is a clean studio-lit model dropped against a daylit city. The cues to fix:

Match the light. If the Photography Style is "golden hour," add a Custom Direction like "warm rim light from camera-left, soft shadow falling camera-right" so the model's lighting agrees with the sun position implied by the backdrop.
Specify depth of field. "Shallow depth of field, background in soft focus" for editorial, "deep focus, sharp street detail" for streetwear and campaign work.
Specify the surface. "Wet asphalt with curb reflection," "limestone cobblestone," "salt-streaked sidewalk." Contact shadows have more to grip when the ground is described.
Specify density. "Quiet morning street, two pedestrians in soft focus" reads differently from "rush-hour midtown, blurred motion of taxis." Pick one.
Pick a time of day in plain words. "Blue hour, faint sodium street lights," "overcast midday, soft even shadows," "night, neon and shop-window light, wet pavement reflections." Time-of-day phrasing carries more weight than abstract mood words.

These layer onto a normal on-model workflow. The Fashion Model stays consistent across the campaign, the Photography Style holds the photographic feel, the Background holds the place, and Custom Directions carry the specifics that turn "a street" into "this street."