|
|
--- |
|
|
tags: |
|
|
- text-to-image |
|
|
- lora |
|
|
- diffusers |
|
|
- template:diffusion-lora |
|
|
base_model: black-forest-labs/FLUX.1-Kontext-dev |
|
|
instance_prompt: >- |
|
|
[photo content], render the image from the left-side perspective, keeping |
|
|
consistent lighting, textures, and proportions. Maintain the realism of all |
|
|
surrounding elements while revealing previously unseen left-side details |
|
|
consistent with the object’s or scene’s structure. |
|
|
license: other |
|
|
license_name: flux-1-dev-non-commercial-license |
|
|
license_link: LICENSE.md |
|
|
language: |
|
|
- en |
|
|
pipeline_tag: image-to-image |
|
|
library_name: diffusers |
|
|
--- |
|
|
|
|
|
 |
|
|
|
|
|
# **Kontext-CAM-Left-View** |
|
|
|
|
|
The Kontext-CAM-Left-View is an experimental adapter for black-forest-lab's FLUX.1-Kontext-dev, designed to generate a left-side perspective of the scene while preserving consistent lighting, textures, and proportions. The model maintains the realism of all surrounding elements and accurately reveals previously unseen left-side details, ensuring seamless perspective alignment and environmental coherence. It was trained on 800 image pairs (400 start images and 400 end images) to deliver high-fidelity, geometry-consistent left-side viewpoint generation. |
|
|
|
|
|
> [!note] |
|
|
[photo content], render the image from the left-side perspective, keeping consistent lighting, textures, and proportions. Maintain the realism of all surrounding elements while revealing previously unseen left-side details consistent with the object’s or scene’s structure. |
|
|
|
|
|
> You modified the prompt, altering its properties and subjective elements. Note: this is an experimental adapter and may contain artifacts. |
|
|
|
|
|
--- |
|
|
|
|
|
## **Sample Inferences : Demo** |
|
|
|
|
|
<table style="width:100%; border-collapse:collapse;"> |
|
|
<tr> |
|
|
<td style="width:50%; text-align:center;"> |
|
|
<img src="https://cdn-uploads.huggingface.co/production/uploads/65bb837dbfb878f46c77de4c/lZ8asnkoamFUH1ClFgn6H.jpeg" |
|
|
alt="Kontext-CAM-Left-View" style="width:100%; height:auto;"/> |
|
|
</td> |
|
|
<td style="width:50%; text-align:center;"> |
|
|
<img src="https://cdn-uploads.huggingface.co/production/uploads/65bb837dbfb878f46c77de4c/F92WuRNLReDYS-nXXBLUz.webp" |
|
|
alt="Kontext-CAM-Left-View" style="width:100%; height:auto;"/> |
|
|
</td> |
|
|
</tr> |
|
|
</table> |
|
|
|
|
|
<table style="width:100%; border-collapse:collapse;"> |
|
|
<tr> |
|
|
<td style="width:50%; text-align:center;"> |
|
|
<img src="https://cdn-uploads.huggingface.co/production/uploads/65bb837dbfb878f46c77de4c/Txk4Mnk7q6wkGFdpe276J.jpeg" |
|
|
alt="Kontext-CAM-Left-View" style="width:100%; height:auto;"/> |
|
|
</td> |
|
|
<td style="width:50%; text-align:center;"> |
|
|
<img src="https://cdn-uploads.huggingface.co/production/uploads/65bb837dbfb878f46c77de4c/uHdGFGI-4plezer-JSAjs.webp" |
|
|
alt="Kontext-CAM-Left-View" style="width:100%; height:auto;"/> |
|
|
</td> |
|
|
</tr> |
|
|
</table> |
|
|
|
|
|
--- |
|
|
|
|
|
## Parameter Settings |
|
|
|
|
|
| Setting | Value | |
|
|
| ------------------------ | ------------------------ | |
|
|
| Module Type | Adapter | |
|
|
| Base Model | FLUX.1 Kontext Dev - fp8 | |
|
|
| Trigger Words | [photo content], render the image from the left-side perspective, keeping consistent lighting, textures, and proportions. Maintain the realism of all surrounding elements while revealing previously unseen left-side details consistent with the object’s or scene’s structure. | |
|
|
| Image Processing Repeats | 42 | |
|
|
| Epochs | 22 | |
|
|
| Save Every N Epochs | 1 | |
|
|
|
|
|
Labeling: DeepCaption-VLA-7B(natural language & English) |
|
|
|
|
|
Total Images Used for Training : 800 Image Pairs (400 Start, 400 End) |
|
|
|
|
|
## Training Parameters |
|
|
|
|
|
| Setting | Value | |
|
|
| --------------------------- | --------- | |
|
|
| Seed | - | |
|
|
| Clip Skip | - | |
|
|
| Text Encoder LR | 0.00001 | |
|
|
| UNet LR | 0.00005 | |
|
|
| LR Scheduler | constant | |
|
|
| Optimizer | AdamW8bit | |
|
|
| Network Dimension | 64 | |
|
|
| Network Alpha | 32 | |
|
|
| Gradient Accumulation Steps | - | |
|
|
|
|
|
## Label Parameters |
|
|
|
|
|
| Setting | Value | |
|
|
| --------------- | ----- | |
|
|
| Shuffle Caption | - | |
|
|
| Keep N Tokens | - | |
|
|
|
|
|
## Advanced Parameters |
|
|
|
|
|
| Setting | Value | |
|
|
| ------------------------- | ----- | |
|
|
| Noise Offset | 0.03 | |
|
|
| Multires Noise Discount | 0.1 | |
|
|
| Multires Noise Iterations | 10 | |
|
|
| Conv Dimension | - | |
|
|
| Conv Alpha | - | |
|
|
| Batch Size | - | |
|
|
| Steps | 3300 & 400(warm up) | |
|
|
| Sampler | euler | |
|
|
|
|
|
--- |
|
|
|
|
|
## Trigger words |
|
|
|
|
|
You should use `[photo content]` to trigger the image generation. |
|
|
|
|
|
You should use `render the image from the left-side perspective` to trigger the image generation. |
|
|
|
|
|
You should use `keeping consistent lighting` to trigger the image generation. |
|
|
|
|
|
You should use `textures` to trigger the image generation. |
|
|
|
|
|
You should use `and proportions. Maintain the realism of all surrounding elements while revealing previously unseen left-side details consistent with the object’s or scene’s structure.` to trigger the image generation. |
|
|
|
|
|
## Download model |
|
|
|
|
|
[Download](/prithivMLmods/Kontext-CAM-Left-View/tree/main) them in the Files & versions tab. |
