> ## Documentation Index
> Fetch the complete documentation index at: https://docs.repli5.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Variation Generator

> Generate realistic material, color, and lighting variations of close-up object photos while preserving existing surface defects and wear.

## Overview

The **Variation Generator** on **the Repli5 platform** takes close-up RGB images of an object (either a dataset or a single uploaded image) and transforms their appearance—such as surface color, material finish, or ambient lighting—based on a natural-language prompt.

> [!NOTE]
>
> **Defects & Geometry Are Preserved:** Using edge-guided conditioning, the model preserves the exact object geometry as well as existing dents, corner wear, scratches, and surface defects while applying your requested variation.

***

## Step-by-Step Guide

**1. 1. Upload / Dataset — Choose Your Input**

In Card **1 (Upload / Dataset)**, choose whether to run on a saved dataset or a single uploaded photo:

* **Dataset tab *(Default)*:** Select a dataset from your library or active project. You can select individual images on the page, **Select all on page**, or run across the **entire dataset**.
* **Upload tab:** Click **Upload image** to upload a single close-up RGB image directly from your computer.

**2. 2. Variation Templates & Prompt — Describe the Change**

In Card **2 (Variation)**:

* **Saved Templates (`Template Bar`):** Load an existing variation template from your active project or personal templates, or save your current prompt as a reusable template for future runs.
* **Variation Prompt:** Describe the visual change you want applied to the object or background.
  * *Example:* `"Change the color of the phone to light gray"`
  * *Example:* `"Change the metal bracket finish to brushed anodized black"`
  * *Example:* `"Change the studio lighting to warm directional side-lighting"`
* **Enhance Button:** Click **Enhance** below the text box to have the AI expand a brief variation request into a rich, structured prompt.
* **Randomized Choice Groups:** Use flat curly-brace choices such as `change the color of the housing to {matte light gray | anodized black | brushed silver}` to pick one option at random for each image in a batch.

> [!TIP]
>
> You do not need to describe the existing scratches or dents in your prompt—the pipeline automatically locks the structural edges and surface wear from the source image.

**3. 3. Summary, Evaluation Rubric & Launch**

In Card **3 (Summary)**:

1. Verify your **Input** count and **Variation** prompt.
2. Check the **Required credits** (`1 credit` per generated image) against your available balance.
3. *(Optional)* Select an **Evaluation Rubric Set** (defaults to the *Defect variation* rubric, which scores `Preserves original defect`).
4. Click **Augment** to enqueue the job.

***

## Reviewing & Exporting Results

Once submitted, your job appears in the **Augment History** panel at the bottom of the page:

* **Quality Evaluation:** Each generated image is automatically scored on whether it preserves the original defect while applying the requested variation (see [Quality Evaluation & Review](https://docs.repli5.com/evaluation/quality-evaluation.md)).
* **Inspect, Approve / Reject & Regenerate:** Compare `rgb_raw` vs. `rgb_out` side-by-side, mark outputs as **Approved** or **Rejected**, or click **Regenerate** on any image to replace it in-place.
* **Download:** Completed batch runs automatically create an output dataset in your **Data Library**, ready for one-click ZIP download.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.