| --- |
| language: |
| - en |
| license: cc-by-nc-nd-4.0 |
| tags: |
| - image-to-image |
| - 16bit-reconstruction |
| - hdr |
| - custom-vision |
| datasets: |
| - custom-dataset-8bit-to-16bit |
| metrics: |
| - name: Median MAE |
| type: regression |
| value: 470 |
| - name: Weighted Mean MAE |
| type: regression |
| value: 747 |
| --- |
| |
| # 16bit-from-8bit Image Reconstruction Model |
|
|
| This model reconstructs **16‑bit per channel images** from standard 8‑bit input images. It is trained on paired 8‑bit and 16‑bit data and optimized to preserve color fidelity and high‑frequency detail. |
|
|
| |
| ## Dataset |
| |
| - **Total images:** 54,580 |
| - **RAW patch images:** 46,000 (~10 GB) |
| - **48‑bit synthetic images:** 8,580 (~2 GB) |
|
|
| - Edge Aware has hand selected 1,975 patches. |
|
|
| ## Evaluation Summary |
|
|
| | MAE is increased from base model, however edge aware training increased sharpness in many cases. Both Entropy and Edge Density has incresed vs. simply scaling 8bit images to 16bit. |
|
|
| ## Intended Use |
|
|
| **Primary Use Cases** |
| - Reconstruction of 16‑bit per channel images from 8‑bit input. |
| - HDR post‑processing and enhancement. |
| - Archival and art restoration workflows. |
|
|
| **Not Intended For** |
| - Lossless scientific measurement or precision tasks. |
|
|
| --- |
