Tool #5 · What we install in your business · In client work

AI Photo Relighting & OptimizationAuthentic Photo Cleanup

The client’s own shop, team, and event photos relit to look professional—with logos, printed signs, and faces verified unchanged.

Last reviewed: 2026-10-22 · Status: In client work

Book a 15-minute fit call Back to all 12 tools in The NexAI Stack
Full stack receipt#5 · In client work · reviewed 2026-10-22
#5AI Photo Relighting & Optimization

Photo cleanup with an OpenAI image model (authentic-photo-cleanup)

In client work
1. What we use it for: We run our `authentic-photo-cleanup` workflow with an OpenAI image model to relight and clean up a client’s real photographs while verifying that every human face, shirt logo, and printed sign stays authentic.
2. Client use case · A print and signage companyState: delivered
Problem

The client provided 590 raw shop and installation photos (deduplicated to 510), several shot under harsh shop lighting, and fake AI stock photos would undermine credibility for a custom signage company.

What the tool did

Used an OpenAI image model (`gpt-image-2.5-sunburst` via our `authentic-photo-cleanup` skill) to relight 4 of the client’s real photos, ran a separate verification check that logos, printed text, and faces survived untouched, and stripped GPS metadata via `sharp`.

Result

Delivered and live on the website (commit `20d10b5`), with 510 deduplicated, WebP/AVIF-optimized photos organized and handed over in Google Drive.

2. Client use case · A members' dinner clubState: live
Problem

A real candlelit dinner-toast photo for the About page had dark, uneven lighting, and synthetic AI people look immediately out of place on a social club site.

What the tool did

AI-relit the club’s real dinner-toast photo (`public/assets/club/about-dinner-toast.webp`) and composited the original printed place cards and member faces back into the final frame.

Result

Live on the About page (`200 OK`) with the real members’ faces and table place cards preserved.

3. What it costs us or the client

~$0.04 to $0.19 per relit image via the OpenAI Image API, plus $0 for local `sharp` WebP/AVIF optimization and EXIF GPS stripping (as of 2026-10-22).

4. Setup notes (how it is wired in)

Takes the original client photo, runs a lighting-only pass through the OpenAI image model (`authentic-photo-cleanup` skill), composites original faces or small typography back over the relit image when needed, verifies logos/faces against the source photo, and outputs responsive WebP/AVIF via `scripts/optimize-images.mjs`.

5. What broke & how it was handled

Unconstrained image-to-image passes subtly warp small printed text, shirt logos, and facial features. On the members’ dinner club toast photo, a full pass softened the place cards and faces, so we had to composite the original faces and place cards back over the relit background; earlier we also generated a temporary home-page stand-in with Google Gemini (commit `0862a34`) that was later replaced because synthetic stand-ins never match real client photography.

6. What we compared it against & why we picked it

Compared against pure synthetic image generation with Google Gemini (used for a temporary stand-in on the dinner club site and for scene generation inside our own pFrame.ai SaaS alongside Photoroom, Pixelbin, and BiRefNet). For service-business websites, we picked `authentic-photo-cleanup` on real client photos because buyers spot fake AI stock imagery immediately.

7. Last reviewed: 2026-10-228. Video / tutorial: Planned build-with-me tutorial: "Relighting real client photos with AI without warping faces or logos"
All stack tools
← #4: CallRail + GTM + Search Console#6: Stripe →
How an engagement runsthree steps, in order

Three separate steps. Stop after any of them.

Step 1 is a free 15-minute fit call. Step 2 is the AI Staff Assessment, $500, a written assessment you keep. Step 3 is your first hire at a published price, or you take the assessment and stop.

Step 1

Fit call

$0 · 15 min
  • tell me which job eats your week: the phone, the inbox, marketing or support
  • I tell you whether an AI hire fits, and whether I am the right person for it
  • you leave with the first hire I would start with
  • a slide deck

video call · same week

book it
Step 2

AI Staff Assessment

$500
  • a 60-minute video call about how your business runs
  • my written assessment the same day: whether an AI hire would pay for itself, the job I would hand off first, what to settle first, and the fixed price for that first hire
  • not taken off the build, and no separate quoting call: the assessment names the package at its published price
  • proposals written before anyone learned your business

paid when you book · not taken off the build

start here
Step 3

Your first hire

fixed price
  • Front Desk Hire or Brain + First Hire, at the published price
  • built in 2 to 6 weeks, then 30 days of tuning
  • or take the assessment and stop there
  • bench developers you never met

priced in the assessment

see the prices

Want Authentic Photo Cleanup wired into your business?

Bring the workflow or codebase to a free 15-minute fit call. We will tell you how we wire Authentic Photo Cleanup in production, what breaks first, and whether a $500 AI Staff Assessment fits your scope.

Book a 15-minute fit call
or email david@nexaiadvisors.com about Authentic Photo Cleanup