Images — examples
Each example is a real job, built and run on Madoo. The pattern is the one of the whole section: the AI invents what no photo contains; the deterministic steps make exact what must be exact — the white of a marketplace, the size of each channel, the brand’s mark and fonts, a code that must scan, a rule that must hold. Every example shows its results and its real cost, explains its steps, and installs into your workspace with an API key.
| Example | What it shows | Credits per run |
|---|---|---|
| I01 — One product photo, a launch kit for every channel | A perfume bottle’s photo becomes a marketplace image checked against the rules, a signed studio shot, an Instagram post in the brand’s fonts, a story whose text GPT Image 2 writes, a surprise banner and a contact sheet; the engine is a Sub-workflow that made a second product’s kit unchanged | ≈26 per product |
| I02 — An empty room, staged, at dusk, and listed | An empty room furnished while the photograph stays exactly as taken, the same room at dusk, and the listing pack: photo with headline and price, before-and-after, a flyer with a QR code read back before delivery, a social post | 10 |
| I03 — Four press photos, a carousel ready to post | A launch’s press photos become an Instagram carousel: a poster cover whose title GPT Image 2 writes, four slides with one house look, numbered and signed, and the words — alt texts, captions in two languages, the post, hashtags — read from the pictures as JSON | 10 |
| I04 — Three garments, one model, one set: a lookbook with shop codes | Flat garments worn by one model, each styled as said, photographed in one set made for the collection as in a real shoot there; cards with name, price and a QR code to the product, read back; the lookbook page | 20 |
| I05 — Your pet as an old-master portrait | A dog’s photo painted as a seventeenth-century noble in the manner of a Velázquez, the dog kept exactly; a 30 × 40 cm CMYK print with a plaque, a framed post, a wallpaper and a sticker | 11 |
| I06 — A name, three personalised gifts and the proof | A customer’s name set exactly on three blank gifts and rendered as embroidery, gold foil and wood burning; the proof for the customer, the order in JSON for the workshop | 15 |
| I07 — A hotel’s photos screened for a booking portal | Eight photos measured — size, exposure, sharpness, people in the picture —, the three that pass cut to 1280 × 720, a line with the numbers and the reason for every photo; the verdicts an AI judgement gave on the same photos, without a model | 0 |
| I08 — A team page from whatever photos people send | Six photos taken every way — and a logo — become the same square head shot for everyone, framed by Face Detect, with the tone measured inside each face; the page from a template, and the list of portraits to ask for | 0 |
| I09 — Two generated packshots checked against the real product | A studio packshot and a warm mood shot made by Product Photography; the jacket alone, cut out of each, measured against the real one — colour shift, light change, texture kept — and only the faithful picture published | 13 |
| I10 — A catalogue in the brand colours | Three products in a cosy autumn scene, each measured against the brand colours by Color Palette and made again with the measured violations as the correction, until it is on brand; one Repeat Until Accepted per product, in one run | ≈ 40 (45 at most) |
| I12 — The same model across a campaign | A coat campaign of three scenes with the brand’s model, a character made with Madoo: each picture’s face, found by Locate Subject from her portrait and role, rated against her portrait on features, skin and hair, and made again with what differed until it is her | ≈ 15 (≈ 50 at most) |
| I11 — A product placed in a scene, faithful to the real one | A sixteenth-century pharmacy jar on an old pharmacy counter, cropped by Locate Subject and judged by Decide with a Model against the museum photo on what the viewpoint cannot explain, corrected by Image Analysis told the verdict, until it is faithful | ≈ 5 (17 at most) |
To install an example, see Installing an example: the same
installer works for these workflows (--from the documentation site, or a local folder).