---
title: voir-mage-pilot48-albedo-v1
canonical_url: "https://www.modelscope.cn/models/ApacheOne/voir-mage-pilot48-albedo-v1"
md_url: "https://www.modelscope.cn/models/ApacheOne/voir-mage-pilot48-albedo-v1.md"
repository: ApacheOne/voir-mage-pilot48-albedo-v1
last_updated: 2026-07-28
license: mit
pipeline_tag: image-to-image
tasks:
  - image-to-image
library_name:
  - pytorch
frameworks:
  - pytorch
downloads: 2
stars: 0
tags:
  - albedo
  - mage-flow
  - reservoir-computing
  - intrinsic-images
  - pytorch
---

# voir-mage-pilot48-albedo-v1

> voir-mage-pilot48-albedo-v1 - ApacheOne 在 ModelScope 开源的模型。VOIR Real Mage Albedo v1

ApacheOne/voir-mage-pilot48-albedo-v1 是 ModelScope 魔搭社区上的image-to-image模型，采用 mit 许可。

- **Repository**: ApacheOne/voir-mage-pilot48-albedo-v1
- **License**: mit
- **Tasks**: image-to-image
- **Tags**: albedo, mage-flow, reservoir-computing, intrinsic-images, pytorch
- **Downloads**: 2
- **Stars**: 0
- **Last updated**: 2026-07-28

Source: https://www.modelscope.cn/models/ApacheOne/voir-mage-pilot48-albedo-v1

---

# VOIR Real Mage Albedo v1

Full-resolution trainable albedo decoder using frozen internal states from
`microsoft/Mage-Flow-Edit-Turbo`.

## Training data

- Cache dataset: `ApacheOne/voir-mage-pilot48-cache-v1`
- Actual Mage cache images: 38 train / 10 validation
- Projected hidden channels: 2240
- Sampler channels: 2688
- Auxiliary channels: 63

## Best held-out validation

```json
{
  "epoch": 26,
  "train_loss": 0.08465919854413521,
  "lr": 0.00040548211460896867,
  "seconds": 299.20014929771423,
  "validation": {
    "loss": 0.0830109141767025,
    "rgb": 0.038485938543453814,
    "gradient": 0.024017343949526548,
    "ssim": 0.15509827733039855,
    "chroma": 0.02399692116305232,
    "color": 0.024400425766361877,
    "multiscale": 0.03854453098028898,
    "total": 0.0830109141767025,
    "mae": 0.03898913008160889,
    "mse": 0.007759020489174873,
    "psnr": 23.279880106718778,
    "ssim_7x7": 0.8449017226696014,
    "global_ssim": 0.852774053812027
  }
}
```

The training set contains only twelve real aerial-image pairs. This checkpoint is
a real Mage-conditioned model, but it is an initial narrow-domain model rather
than a general-purpose albedo estimator. More paired scenes are required for broad
generalization.