nodes/CosmicLaca-ComfyUI_Primere_Nodes
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

V2.0.0 - check #8 - new readme

Browse Source
This commit is contained in:
DESKTOP-TVBJISQ\Primere
2026-06-22 13:39:03 +02:00
parent 2273d5f74d
commit df277bf25a
7 changed files with 1319 additions and 880 deletions
Download Patch File Download Diff File Expand all files Collapse all files
Workflow/Manual/nodes/basic_production_plus_workflow.md
+86 -177
View File
@@ -1,219 +1,128 @@
# <ins>Several style nodes:</ins>
# Basic Production Plus Workflow - Node Groups Manual
### Style pile:
Basic Production Plus workflow contains everything that exists in the Basic Production workflow. Here are only the added features listed. <ins>[Read the Basic Production workflow manual first.](./basic_production_workflow.md)</ins>
<img src="pstyle_pile.jpg" width="400px">
---
A node that enhances your prompts by adding predefined artistic styles and attributes with fine-grained control over their influence.
# <ins>Reference Image Source:</ins>
#### Description:
<img src="reference_example.jpg" width="800px">
This node allows you to enrich your prompts with various artistic styles, concepts, and attributes. You can stack multiple style elements while controlling their individual strengths, making it easy to create complex, nuanced prompts without manual typing.
The Reference Image Source node provides the ability to feed external reference images into editing-capable models such as Qwen-Edit, FireRed, or similar image-to-image architectures. This is essential for workflows that modify existing images based on prompt instructions rather than generating from scratch.
#### Features:
---
- Add multiple style elements to positive and negative prompts
- Fine-tune the strength of each style component
- Categories include:
- Art Types (3d-rendering, digital-artwork, drawing, painting, photo, vector-art)
- Concepts
- Artists
- Art Movements
- Colors
- Directions
- Moods
### Primere Multi Image
#### Style Controls:
**Purpose:** Load and prepare up to 16 source images for editing models. The node handles resizing, padding, batching, and concatenation, outputting a flexible image list that editing models consume.
Each category includes:
- Main selector for choosing style elements
- Strength slider (0.00-1.00) to control influence
---
#### Benefits:
**Inputs:**
- Create consistent style presets
- Quickly experiment with different artistic styles
- Fine-tune style influence without prompt editing
- Stack multiple styles with precise control
- Separate positive and negative style applications
| Input | Type | Purpose |
|-------|------|---------|
| `image` | Required *IMAGE* | Primary image input (always connected) |
| `image_2` – `image_16` | Optional *IMAGE* | Additional source images (up to 16 total) |
| `number_of_images` | *INT* | How many of the 16 slots to process (1–16) |
Example montage. The first pic made without style addition, all pics generated same seed and settings:
---
<img src="montage_spile.jpg" width="600px">
**Settings:**
<hr>
| Setting | Options | Purpose |
|---------|---------|---------|
| `process_list` | ON / OFF | Enable/disable list output for editing models |
| `resize_source` | ON / OFF | Resize all source images to a target megapixel value |
| `resize_source_mpx` | 0.10 – 48.00 MPX | Target resolution when resizing is enabled |
| `padded_list` | ON / OFF | Pad listed images to uniform size (recommended for editing models) |
| `batch_match` | ON / OFF | Pad batched images to uniform size |
| `batch_padding_color` | white / black | Padding color for uniform sizing |
| `concat_resize_mode` | ON / OFF | Resize final concatenated output to target megapixels |
| `concat_mode` | Horizontal / Vertical / Square | Direction for concatenating images |
| `concat_match_size` | ON / OFF | Match all sources to the first image's dimensions |
| `concat_spacing_width` | 0 – 1024 px | Gap between concatenated images |
| `concat_spacing_color` | white/black/red/green/blue | Spacing color |
### Midjourney styles:
---
<img src="pstyle_mj.jpg" width="400px">
**Outputs:**
A specialized node that brings Midjourney-inspired artistic styles and aesthetics to ComfyUI workflows, allowing fine-grained control over multiple art movements and styles.
| Output | Purpose |
|--------|---------|
| `IMAGE_LIST` | Individual image tensors as a list (best for editing models like Qwen-Edit, FireRed) |
| `IMAGE_BATCH` | All images batched into a single tensor |
| `IMAGE_CONCAT` | Images concatenated into a single composite image |
#### Description:
---
This node enhances your prompts with carefully curated artistic styles inspired by Midjourney's signature aesthetics. It offers granular control over various art movements and styles, each with adjustable strength parameters.
**Recommended usage for editing models:**
#### Special Features:
For the best results with image editing models, use the `IMAGE_LIST` output with these settings:
- mj_artists_styles: Curated artist-specific styles
- mj_art_keywords: Special Midjourney-compatible keywords
- Hierarchical style options (e.g., Realism::Gustave Courbet)
- Individual strength controls (0.00-1.00) for each style
```
process_list: ON
resize_source: ON
resize_source_mpx: 1.00 (or match your model's preferred input resolution)
padded_list: ON
```
#### Benefits:
The resized, padded image list ensures consistent dimensions across all source images, which editing architectures like Qwen-Edit and FireRed expect for reliable processing.
- Create Midjourney-like aesthetics in ComfyUI
- Mix multiple art movements with precise control
- Apply artist-specific styles
- Fine-tune style intensity
- Separate positive and negative style applications
---
#### Example Applications:
**Use cases:**
- Batch editing multiple reference images through Qwen-Edit
- Providing before/after style transfer references to FireRed
- Supplying multiple angles or variants to image-conditioned generators
- Creating composite reference sheets for consistent character editing
As shown in the example images:
1. Precision-focused elephant portrait with strong lines and detailed textures
2. Renaissance-style medieval scene with classical composition
3. Blend of realism and romanticism in character portraits
---
Example montage. The first pic made without style addition, all pics generated same seed and settings:
# <ins>Image Comparison Before/After:</ins>
<img src="montage_mj.jpg" width="600px">
The Basic Production Plus workflow includes built-in before/after comparison for the refinement process. The pre-refiner output and the post-refiner output are both preserved in the pipeline, allowing direct visual assessment of what the detailer nodes changed.
#### Usage Tips:
**How it works:**
1. Select primary art movement (e.g., Realism)
2. Choose specific artist or sub-style
3. Adjust strength for desired intensity
4. Stack multiple styles for complex effects
5. Use keywords for additional refinement
- The raw generator output is tapped before entering the refiner detailer blocks (Face, Eye, Mouth, Hands)
- The final refined output is captured after all detailer passes
- Both images are available for side-by-side comparison
- This lets you evaluate whether each refiner pass improved or altered the image as intended
<hr>
**Workflow position:**
### Emotion styles:
`Generator → [Pre-refiner tap] → Refiner Detailers → [Post-refiner tap] → Saver`
<img src="pstyle_emo.jpg" width="400px">
**Use cases:**
- Verify that face detailer improved facial features without introducing artifacts
- Compare eye enhancement results against the original
- Tune refiner settings by observing before/after differences
- Debug which detailer pass caused unexpected changes
A sophisticated node that infuses emotional layers into your images by adding carefully calibrated emotional qualities to prompts.
---
#### Description:
# <ins>Dual Histogram:</ins>
This specialized node enables the addition of complex emotional states and moods to your generations. It offers fine-tuned control over various emotions, from basic feelings to nuanced psychological states, particularly focusing on subtle variations of melancholic moods.
The Basic Production Plus workflow places **two histogram instances** at different points in the pipeline:
#### Core Emotions:
1. **Raw output histogram** — Monitors the image directly after the decoder, before any Rasterix post-processing
2. **Rasterix output histogram** — Monitors the final image after all Rasterix post-processing nodes
- Love: Romantic and affectionate qualities
- Joy: Happiness and positive expressions
- Surprise: Unexpected and wonder-filled moments
- Anger: Intensity and passion
- Sadness: Subtle expressions of sorrow
- Fear: Tension and apprehension
This dual placement gives you full insight into both the generator's native output quality and the effect of your post-processing adjustments.
#### Melancholic Variations:
**Pipeline position:**
Refined control over melancholic moods including:
- Melancholia: Classical melancholic expression
- Saudade: Portuguese concept of longing
- Grief: Deep emotional response
- Han: Korean concept of collective sorrow
- Hiraeth: Welsh nostalgia for home
- Mono no aware: Japanese awareness of impermanence
- Nostalgia: Bittersweet remembrance
- Sehnsucht: German romantic yearning
- Good old days: Reminiscent mood
```
Decoder → [Histogram: raw] → Rasterix stack → [Histogram: post-processed] → Saver
```
#### Benefits:
**Benefits:**
- Compare tonal distribution before and after grading
- Detect clipping introduced during post-processing
- Verify that your Rasterix adjustments improved the image rather than degrading it
- Fine-tune the Rasterix pipeline with objective visual feedback
- Add emotional depth to portraits
- Create mood-specific variations of the same image
- Control the intensity of emotional expression
- Layer multiple emotions for complex feelings
- Separate positive and negative emotional qualities
For detailed histogram settings and behavior, see the [Rasterix Nodes Manual](./rasterix.md).
#### Example Applications:
As shown in the example portraits:
1. Neutral baseline portrait
2. Joy and warmth expression
3. Subtle melancholic undertones
4. Contemplative mood variation
Example montage. The first pic made without style addition, all pics generated same seed and settings:
<img src="montage_emo.jpg" width="600px">
One more montage of Emotion style node:
<img src="montage_emo_1.jpg" width="600px">
#### Usage Tips:
1. Select primary emotion
2. Choose specific variation if available
3. Adjust strength (0.00-1.00) for intensity
4. Layer multiple emotions for complexity
5. Use with portrait-focused generations
<hr>
### Camera lens styles:
<img src="pstyle_lens.jpg" width="400px">
A custom node that enhances prompts with detailed camera lens characteristics for improved image generation results.
#### Features:
- Categorized lens presets by photography type
- Detailed lens descriptions and technical specifications included to prompt
- Adjustable strength parameters
- Manufacturer-specific presets
#### Usage:
1. Add "Primere Lens Styles" node to workflow
2. Connect `opt_pos_style` for positive prompts
3. Connect `opt_neg_style` for negative prompts
4. Select lens category
5. Adjust strength (0.0-1.0)
6. Choose specific lens preset
### Categories:
#### Art:
- Circular Fisheye
- Special Effects
- Tilt-Shift
#### Landscape:
- Wide Angle
- Ultra-Wide
- Panoramic
#### Portrait:
- Art Lens (Sony)
- Bokeh Lens (Sony)
- Lightroom Lens (Fujifilm)
- Low Light Lens (Sigma)
- Macro Lens (Canon)
- Portrait Lens (Nikon)
- Prime Lens (Canon)
- Telephoto Lens (Canon)
- Vintage Lens (Nikon)
- Zoom Lens (Tamron)
#### Additional Categories (with seevral lens type):
- Product
- Sport
- Street
- Wildlife
Example montage. The first pic made without style addition, all pics generated same seed and settings:
<img src="montage_lens.jpg" width="600px">
Example images demonstrate:
- Bokeh effects
- Background compression
- Detail retention
- Subject isolation
- Color rendition
---
Workflow/Manual/nodes/basic_production_workflow.md
+49
View File
@@ -500,3 +500,52 @@ Remember: DiT prompt purification is essential for optimal performance with mode
# <ins>Advanced style block:</ins>
<img src="Basic_style_block.jpg" width="800px">
The Advanced style block extends the Basic workflow's style injection by providing **4 parallel Primere Custom Styles nodes**, each independently configurable with a different `.toml` source from `Toml/Styles/`. This allows composing styles from multiple files simultaneously — for example, applying a mood from `emotions_style.toml` while layering lighting from `light_physics_style.toml` and lens effects from `lens_styles.toml`.
**How it works:**
- Each of the 4 nodes is an independent instance of the same universal style handler described in the [Basic workflow manual](./basic_workflow.md#style-injection-group).
- Each node selects its own `style_source` (`.toml` file) and exposes per-section dynamic widgets for style selection and strength.
- The outputs (`STYLE+`, `STYLE-`) can be mixed, chained, or routed individually to the prompt encoder.
- You can add more instances by duplicating a node — there is no hard limit.
**Use cases:**
- Combine artistic movement reference + mood + camera lens profile in a single generation
- Keep dedicated TOML files for different projects or clients and toggle between them per node
- A/B test style combinations by enabling/disabling individual nodes
For a detailed description of the Primere Custom Styles node, see the [Style Injection Group](./basic_workflow.md#style-injection-group) in the Basic workflow manual.
---
# <ins>Advanced Rasterix block:</ins>
<img src="Rasterix_group_small.jpg" width="800px">
The Advanced Rasterix block provides a comprehensive set of modular image post-processing nodes for professional-grade finishing. Each node operates independently with its own `use_*` enable switch, allowing you to build a custom processing pipeline.
**Available Rasterix nodes in this block:**
- **Selective Tone** — Per-zone exposure control (highlights, midtones, shadows, blacks)
- **Brightness Contrast** — Global brightness/contrast with adaptive mode
- **Color Balance** — Channel-based grading per tonal range
- **Grain** — Film grain with tonal distribution and color tinting
- **Histogram** — Luminance and color distribution visualization
- **Auto Levels** — Automatic tonal range optimization
- **Film Rendering** — Cinematic color grading and film simulation
- **Levels** — Manual shadow/midtone/highlight adjustment
- **DxO Style** — DxO-inspired lens and color correction profiles
- **Clipping Display** — Visual feedback for clipped highlights/shadows
- **Tone Lab** — Advanced curve-based tonal manipulation
**Pipeline position:**
`Sampler → Decoder → Rasterix post-processing stack → Histogram → Saver`
Nodes can be chained in any order. Use the `use_*` switch on each node to enable/disable processing without removing it from the pipeline.
> **For the complete Rasterix documentation including detailed settings, behavior, and use cases, see the [Rasterix Nodes Manual](./rasterix.md).**
---
Workflow/Manual/nodes/basic_workflow.md
+53 -39
View File
@@ -237,58 +237,72 @@ Pipeline position:
---
The Style Injection group adds professional-grade art style control to the Basic workflow. It enables rapid selection and injection of complex artistic styles, concepts, artists, movements, colors, directions, and moods directly into your prompts. This node works in tandem with the 12-prompt selector system for efficient style experimentation and iteration.
The Style Injection group adds professional-grade art style control to the Basic workflow. It enables rapid selection and injection of complex artistic styles, concepts, moods, lighting, and more directly into your prompts. This node works in tandem with the 12-prompt selector system for efficient style experimentation and iteration.
<img src="./Basic_style_node.jpg" width="300px">
---
### Primiere Style Pile
### Primere Custom Styles
Purpose: Multi-category style composer that builds precise positive and negative style descriptors from curated preset lists. All combo options are dynamically loaded from the external `stylepile.toml` file, allowing easy customization and extension of available styles.
**Purpose:** Universal TOML-based style handler that loads all style definitions from external `.toml` files. The node scans the `Toml/Styles/` directory and presents every `.toml` file as a selectable style source. Each section within the file becomes a pair of dynamic widgets — a dropdown and a strength slider — giving you full control over style composition.
**Key benefit:** Users can create custom `.toml` files with their own style sections and drop them into `Toml/Styles/`. The node automatically detects new files and loads their contents, with no code changes or configuration needed. There is no limit to the number of custom style sources.
The node ships with 18 example `.toml` files covering emotions, lighting, lens effects, art movements, organic forms, perception, and more.
---
Outputs:
**How It Works:**
| Output | Purpose |
| --------------- | ----------------------------------------------------------------------- |
| `opt_pos_style` | STYLE+ : Positive style injection string (ready to connect to prompt) |
| `opt_neg_style` | STYLE- : Negative style injection string (ready to connect to prompt) |
1. On node load, `Nodes/Styles.py` scans `Toml/Styles/*.toml` and populates the `style_source` dropdown.
2. When you select a source file, the frontend fetches its parsed sections and creates dynamic widgets: a **combo dropdown** per section for style selection, and a **number widget** for per-style strength (0–10, step 0.01).
3. On execution, the backend reads your selections via workflow metadata, builds style strings, and merges them with optional manual overrides.
---
Settings:
**Inputs:**
| Setting | Purpose |
| -------------------------- | ----------------------------------------------------------------------- |
| `art-type` | Base art format (painting, photo, digital-artwork, 3d-rendering, etc.) |
| `art-type_strength` | Strength multiplier for the selected art-type |
| `concepts` | High-level conceptual qualifiers (Extreme, Masterpiece, Cinematic, etc.)|
| `concepts_strength` | Strength multiplier for concepts |
| `artists` | Artist/illustrator reference |
| `artists_strength` | Strength multiplier for artist influence |
| `art-movements` | Historical/modern art movement (Action Painting, Surrealism, etc.) |
| `art-movements_strength` | Strength multiplier for art movement |
| `colors` | Color palette or processing style (CMYK Colors, Vivid Colors, etc.) |
| `colors_strength` | Strength multiplier for color treatment |
| `directions` | Rendering direction/quality descriptor (Masterpiece, Realistic, etc.) |
| `directions_strength` | Strength multiplier for direction |
| `moods` | Emotional or atmospheric tone (Energetic, Romantic, Dramatic, etc.) |
| `moods_strength` | Strength multiplier for mood |
| Input | Type | Purpose |
|-------|------|---------|
| `style_source` | Required *COMBO* | Select a `.toml` file from `Toml/Styles/` |
| `opt_pos_style` | Optional *STRING* | Manual positive style override text |
| `opt_neg_style` | Optional *STRING* | Manual negative style override text |
| *(dynamic)* | *COMBO + FLOAT* | Per-section dropdown + strength, created per TOML section |
#### Behavior:
* Each selected category appends the corresponding Positive/Negative text defined in `stylepile.toml`.
* Strength values are automatically applied as weighted prompt syntax `(selected_item:strength)`.
* Empty or default selections are safely skipped.
* The node outputs clean, ready-to-use strings that can be injected into any prompt input without manual editing.
* Works with the existing prompt encoder, dynamic prompts, and the 12-prompt selector system.
---
#### Use Cases:
* One-click style switching during prompt development (e.g. painting → photo → vector-art)
* Artist-specific or movement-specific testing
* Layering multiple style elements with precise strength control
* Consistent aesthetic application across batch generations or A/B testing
**Outputs:**
| Output | Purpose |
|--------|---------|
| `STYLE+` | Positive style injection string (ready to connect to prompt) |
| `STYLE-` | Negative style injection string (ready to connect to prompt) |
---
**Adding Custom Style Sources:**
Create a new `.toml` file in `Toml/Styles/` with this structure:
```toml
[SectionName]
[SectionName.N_0]
MainStyle = "Your Style"
SubStyle = ""
Positive = "Positive prompt text for this style"
Negative = "Negative prompt text for this style"
```
Each top-level section becomes a widget pair. The `Positive` field feeds into `STYLE+`, and the `Negative` field into `STYLE-`. You can add as many sections and files as needed — the node will load all of them automatically.
---
**Use Cases:**
* One-click style switching during prompt development
* Custom style libraries shared across projects
* Themed style packs for specific genres or aesthetics
* Rapid A/B testing of different artistic directions
---
@@ -296,9 +310,9 @@ Settings:
Pipeline position:
`Prompt Sources / 12-Prompt Selector → Primiere Style Pile (STYLE+ / STYLE-) → Prompt Encoder → Sampler`
`Prompt Sources / 12-Prompt Selector → Primere Custom Styles (STYLE+ / STYLE-) → Prompt Encoder → Sampler`
The Style Pile node sits early in the prompt pipeline and feeds directly into the central prompt builder. It is fully compatible with CSV/TOML prompt readers and the existing style injection points.
The style injection sits early in the prompt pipeline and feeds directly into the prompt encoder. It is fully compatible with CSV/TOML prompt readers, dynamic prompts, and the existing style injection points.
---
@@ -354,4 +368,4 @@ Pipeline position:
`Prompt Development Group → Style Injection Group → Prompt Encoder → Sampler`
The selected prompt channel flows directly into the Primiere Style Pile (STYLE+ / STYLE-) for seamless combination of your base prompt with artistic styling.
The selected prompt channel flows directly into the Primere Custom Styles node (STYLE+ / STYLE-) for seamless combination of your base prompt with artistic styling.
Workflow/Manual/nodes/reference_example.jpg
BIN
View File
Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 325 KiB

Workflow/Manual/nodes/style_examples.md
+219
View File
@@ -0,0 +1,219 @@
# <ins>Several style nodes:</ins>
### Style pile:
<img src="pstyle_pile.jpg" width="400px">
A node that enhances your prompts by adding predefined artistic styles and attributes with fine-grained control over their influence.
#### Description:
This node allows you to enrich your prompts with various artistic styles, concepts, and attributes. You can stack multiple style elements while controlling their individual strengths, making it easy to create complex, nuanced prompts without manual typing.
#### Features:
- Add multiple style elements to positive and negative prompts
- Fine-tune the strength of each style component
- Categories include:
- Art Types (3d-rendering, digital-artwork, drawing, painting, photo, vector-art)
- Concepts
- Artists
- Art Movements
- Colors
- Directions
- Moods
#### Style Controls:
Each category includes:
- Main selector for choosing style elements
- Strength slider (0.00-1.00) to control influence
#### Benefits:
- Create consistent style presets
- Quickly experiment with different artistic styles
- Fine-tune style influence without prompt editing
- Stack multiple styles with precise control
- Separate positive and negative style applications
Example montage. The first pic made without style addition, all pics generated same seed and settings:
<img src="montage_spile.jpg" width="600px">
<hr>
### Midjourney styles:
<img src="pstyle_mj.jpg" width="400px">
A specialized node that brings Midjourney-inspired artistic styles and aesthetics to ComfyUI workflows, allowing fine-grained control over multiple art movements and styles.
#### Description:
This node enhances your prompts with carefully curated artistic styles inspired by Midjourney's signature aesthetics. It offers granular control over various art movements and styles, each with adjustable strength parameters.
#### Special Features:
- mj_artists_styles: Curated artist-specific styles
- mj_art_keywords: Special Midjourney-compatible keywords
- Hierarchical style options (e.g., Realism::Gustave Courbet)
- Individual strength controls (0.00-1.00) for each style
#### Benefits:
- Create Midjourney-like aesthetics in ComfyUI
- Mix multiple art movements with precise control
- Apply artist-specific styles
- Fine-tune style intensity
- Separate positive and negative style applications
#### Example Applications:
As shown in the example images:
1. Precision-focused elephant portrait with strong lines and detailed textures
2. Renaissance-style medieval scene with classical composition
3. Blend of realism and romanticism in character portraits
Example montage. The first pic made without style addition, all pics generated same seed and settings:
<img src="montage_mj.jpg" width="600px">
#### Usage Tips:
1. Select primary art movement (e.g., Realism)
2. Choose specific artist or sub-style
3. Adjust strength for desired intensity
4. Stack multiple styles for complex effects
5. Use keywords for additional refinement
<hr>
### Emotion styles:
<img src="pstyle_emo.jpg" width="400px">
A sophisticated node that infuses emotional layers into your images by adding carefully calibrated emotional qualities to prompts.
#### Description:
This specialized node enables the addition of complex emotional states and moods to your generations. It offers fine-tuned control over various emotions, from basic feelings to nuanced psychological states, particularly focusing on subtle variations of melancholic moods.
#### Core Emotions:
- Love: Romantic and affectionate qualities
- Joy: Happiness and positive expressions
- Surprise: Unexpected and wonder-filled moments
- Anger: Intensity and passion
- Sadness: Subtle expressions of sorrow
- Fear: Tension and apprehension
#### Melancholic Variations:
Refined control over melancholic moods including:
- Melancholia: Classical melancholic expression
- Saudade: Portuguese concept of longing
- Grief: Deep emotional response
- Han: Korean concept of collective sorrow
- Hiraeth: Welsh nostalgia for home
- Mono no aware: Japanese awareness of impermanence
- Nostalgia: Bittersweet remembrance
- Sehnsucht: German romantic yearning
- Good old days: Reminiscent mood
#### Benefits:
- Add emotional depth to portraits
- Create mood-specific variations of the same image
- Control the intensity of emotional expression
- Layer multiple emotions for complex feelings
- Separate positive and negative emotional qualities
#### Example Applications:
As shown in the example portraits:
1. Neutral baseline portrait
2. Joy and warmth expression
3. Subtle melancholic undertones
4. Contemplative mood variation
Example montage. The first pic made without style addition, all pics generated same seed and settings:
<img src="montage_emo.jpg" width="600px">
One more montage of Emotion style node:
<img src="montage_emo_1.jpg" width="600px">
#### Usage Tips:
1. Select primary emotion
2. Choose specific variation if available
3. Adjust strength (0.00-1.00) for intensity
4. Layer multiple emotions for complexity
5. Use with portrait-focused generations
<hr>
### Camera lens styles:
<img src="pstyle_lens.jpg" width="400px">
A custom node that enhances prompts with detailed camera lens characteristics for improved image generation results.
#### Features:
- Categorized lens presets by photography type
- Detailed lens descriptions and technical specifications included to prompt
- Adjustable strength parameters
- Manufacturer-specific presets
#### Usage:
1. Add "Primere Lens Styles" node to workflow
2. Connect `opt_pos_style` for positive prompts
3. Connect `opt_neg_style` for negative prompts
4. Select lens category
5. Adjust strength (0.0-1.0)
6. Choose specific lens preset
### Categories:
#### Art:
- Circular Fisheye
- Special Effects
- Tilt-Shift
#### Landscape:
- Wide Angle
- Ultra-Wide
- Panoramic
#### Portrait:
- Art Lens (Sony)
- Bokeh Lens (Sony)
- Lightroom Lens (Fujifilm)
- Low Light Lens (Sigma)
- Macro Lens (Canon)
- Portrait Lens (Nikon)
- Prime Lens (Canon)
- Telephoto Lens (Canon)
- Vintage Lens (Nikon)
- Zoom Lens (Tamron)
#### Additional Categories (with seevral lens type):
- Product
- Sport
- Street
- Wildlife
Example montage. The first pic made without style addition, all pics generated same seed and settings:
<img src="montage_lens.jpg" width="600px">
Example images demonstrate:
- Bokeh effects
- Background compression
- Detail retention
- Subject isolation
- Color rendition
Workflow/NewFE/Primere_basic_production_plus_workflow.json
+613 -391
View File
File diff suppressed because it is too large Load Diff
Workflow/NewFE/Primere_basic_production_workflow.json
+299 -273
View File
@@ -1292,7 +1292,7 @@
88
],
"flags": {},
"order": 29,
"order": 28,
"mode": 0,
"inputs": [
{
@@ -2764,7 +2764,7 @@
46
],
"flags": {},
"order": 43,
"order": 42,
"mode": 0,
"inputs": [
{
@@ -2820,7 +2820,7 @@
46
],
"flags": {},
"order": 45,
"order": 44,
"mode": 0,
"inputs": [
{
@@ -3205,7 +3205,7 @@
88
],
"flags": {},
"order": 27,
"order": 26,
"mode": 0,
"inputs": [
{
@@ -4762,7 +4762,7 @@
1807.3454771346862
],
"flags": {},
"order": 34,
"order": 33,
"mode": 0,
"inputs": [
{
@@ -4793,7 +4793,7 @@
154
],
"flags": {},
"order": 28,
"order": 27,
"mode": 4,
"inputs": [
{
@@ -5549,7 +5549,7 @@
138
],
"flags": {},
"order": 36,
"order": 35,
"mode": 4,
"inputs": [
{
@@ -6533,250 +6533,6 @@
"color": "#549494",
"bgcolor": "#408080"
},
{
"id": 548,
"type": "PrimereCustomStyles",
"pos": [
3956.4811553153468,
1359.5925719455224
],
"size": [
340,
414
],
"flags": {},
"order": 18,
"mode": 0,
"inputs": [
{
"name": "opt_pos_style",
"shape": 7,
"type": "STRING",
"link": null
},
{
"name": "opt_neg_style",
"shape": 7,
"type": "STRING",
"link": null
}
],
"outputs": [
{
"name": "STYLE+",
"type": "STRING",
"links": [
5742
]
},
{
"name": "STYLE-",
"type": "STRING",
"links": [
5743
]
}
],
"properties": {
"Node name for S&R": "PrimereCustomStyles"
},
"widgets_values": [
"aftermath_style.toml",
"None",
1,
"None",
1,
"None",
1,
"None",
1
],
"color": "#777d2b",
"bgcolor": "#636917"
},
{
"id": 550,
"type": "PrimereCustomStyles",
"pos": [
4674.872527592744,
1360.573856233446
],
"size": [
340,
606
],
"flags": {},
"order": 39,
"mode": 0,
"inputs": [
{
"name": "opt_pos_style",
"shape": 7,
"type": "STRING",
"link": 5740
},
{
"name": "opt_neg_style",
"shape": 7,
"type": "STRING",
"link": 5741
}
],
"outputs": [
{
"name": "STYLE+",
"type": "STRING",
"links": [
5783
]
},
{
"name": "STYLE-",
"type": "STRING",
"links": [
5784
]
}
],
"properties": {
"Node name for S&R": "PrimereCustomStyles"
},
"widgets_values": [
"aftermath_style.toml",
"None",
1,
"None",
1,
"None",
1,
"None",
1
],
"color": "#777d2b",
"bgcolor": "#636917"
},
{
"id": 551,
"type": "PrimereCustomStyles",
"pos": [
3956.3739090390873,
1817.2914742435973
],
"size": [
340,
414
],
"flags": {},
"order": 26,
"mode": 0,
"inputs": [
{
"name": "opt_pos_style",
"shape": 7,
"type": "STRING",
"link": 5742
},
{
"name": "opt_neg_style",
"shape": 7,
"type": "STRING",
"link": 5743
}
],
"outputs": [
{
"name": "STYLE+",
"type": "STRING",
"links": [
5738
]
},
{
"name": "STYLE-",
"type": "STRING",
"links": [
5739
]
}
],
"properties": {
"Node name for S&R": "PrimereCustomStyles"
},
"widgets_values": [
"aftermath_style.toml",
"None",
1,
"None",
1,
"None",
1,
"None",
1
],
"color": "#777d2b",
"bgcolor": "#636917"
},
{
"id": 549,
"type": "PrimereCustomStyles",
"pos": [
4317.311750382618,
1360.0800425843127
],
"size": [
340,
414
],
"flags": {},
"order": 33,
"mode": 0,
"inputs": [
{
"name": "opt_pos_style",
"shape": 7,
"type": "STRING",
"link": 5738
},
{
"name": "opt_neg_style",
"shape": 7,
"type": "STRING",
"link": 5739
}
],
"outputs": [
{
"name": "STYLE+",
"type": "STRING",
"links": [
5740
]
},
{
"name": "STYLE-",
"type": "STRING",
"links": [
5741
]
}
],
"properties": {
"Node name for S&R": "PrimereCustomStyles"
},
"widgets_values": [
"aftermath_style.toml",
"None",
1,
"None",
1,
"None",
1,
"None",
1
],
"color": "#777d2b",
"bgcolor": "#636917"
},
{
"id": 254,
"type": "ReroutePrimitive|pysssss",
@@ -6789,7 +6545,7 @@
26
],
"flags": {},
"order": 31,
"order": 30,
"mode": 0,
"inputs": [
{
@@ -6849,7 +6605,7 @@
310
],
"flags": {},
"order": 41,
"order": 40,
"mode": 0,
"inputs": [
{
@@ -6951,7 +6707,7 @@
2096
],
"flags": {},
"order": 30,
"order": 29,
"mode": 0,
"inputs": [
{
@@ -7113,7 +6869,7 @@
374
],
"flags": {},
"order": 19,
"order": 18,
"mode": 0,
"inputs": [],
"outputs": [
@@ -7203,7 +6959,7 @@
246
],
"flags": {},
"order": 20,
"order": 19,
"mode": 0,
"inputs": [],
"outputs": [
@@ -7583,7 +7339,7 @@
340.7728271484375
],
"flags": {},
"order": 44,
"order": 43,
"mode": 0,
"inputs": [
{
@@ -7627,7 +7383,7 @@
158
],
"flags": {},
"order": 35,
"order": 34,
"mode": 0,
"inputs": [
{
@@ -7727,7 +7483,7 @@
341.1094055175781
],
"flags": {},
"order": 42,
"order": 41,
"mode": 0,
"inputs": [
{
@@ -7771,7 +7527,7 @@
643.1478271484375
],
"flags": {},
"order": 21,
"order": 20,
"mode": 0,
"inputs": [],
"outputs": [
@@ -7861,7 +7617,7 @@
614
],
"flags": {},
"order": 32,
"order": 31,
"mode": 0,
"inputs": [
{
@@ -7960,7 +7716,7 @@
1298
],
"flags": {},
"order": 38,
"order": 37,
"mode": 0,
"inputs": [
{
@@ -8333,7 +8089,7 @@
446
],
"flags": {},
"order": 22,
"order": 21,
"mode": 0,
"inputs": [],
"outputs": [
@@ -8421,7 +8177,7 @@
26
],
"flags": {},
"order": 37,
"order": 36,
"mode": 0,
"inputs": [
{
@@ -8483,7 +8239,7 @@
"flags": {
"collapsed": false
},
"order": 40,
"order": 39,
"mode": 0,
"inputs": [
{
@@ -8655,7 +8411,7 @@
136
],
"flags": {},
"order": 23,
"order": 22,
"mode": 0,
"inputs": [],
"outputs": [
@@ -8715,7 +8471,7 @@
136.3629386579846
],
"flags": {},
"order": 24,
"order": 23,
"mode": 0,
"inputs": [],
"outputs": [
@@ -8762,7 +8518,7 @@
274
],
"flags": {},
"order": 25,
"order": 24,
"mode": 0,
"inputs": [],
"outputs": [
@@ -9093,6 +8849,276 @@
],
"color": "#d31414",
"bgcolor": "#bf0000"
},
{
"id": 551,
"type": "PrimereCustomStyles",
"pos": [
3956.3739090390873,
1817.2914742435973
],
"size": [
340,
606
],
"flags": {},
"order": 32,
"mode": 0,
"inputs": [
{
"name": "opt_pos_style",
"shape": 7,
"type": "STRING",
"link": 5742
},
{
"name": "opt_neg_style",
"shape": 7,
"type": "STRING",
"link": 5743
}
],
"outputs": [
{
"name": "STYLE+",
"type": "STRING",
"links": [
5738
]
},
{
"name": "STYLE-",
"type": "STRING",
"links": [
5739
]
}
],
"properties": {
"Node name for S&R": "PrimereCustomStyles"
},
"widgets_values": [
"mj_styles.toml",
"Abstract::Collage::Kurt Schwitters",
1,
"Art::Deco::Tamara de Lempicka",
1,
"None",
1,
"None",
1,
"Precisionist::Sculpture::Charles Sheeler",
1,
"None",
1,
"None",
1,
"None",
1,
"None",
1,
"None",
1,
"None",
1
],
"color": "#777d2b",
"bgcolor": "#636917"
},
{
"id": 550,
"type": "PrimereCustomStyles",
"pos": [
4674.872527592744,
1360.573856233446
],
"size": [
340,
606
],
"flags": {},
"order": 45,
"mode": 0,
"inputs": [
{
"name": "opt_pos_style",
"shape": 7,
"type": "STRING",
"link": 5740
},
{
"name": "opt_neg_style",
"shape": 7,
"type": "STRING",
"link": 5741
}
],
"outputs": [
{
"name": "STYLE+",
"type": "STRING",
"links": [
5783
]
},
{
"name": "STYLE-",
"type": "STRING",
"links": [
5784
]
}
],
"properties": {
"Node name for S&R": "PrimereCustomStyles"
},
"widgets_values": [
"light_physics_style.toml",
"None",
1,
"Refraction Systems::Index transition",
1,
"None",
1,
"Interference & Diffraction::Diffraction spread",
1,
"None",
1,
"Fluorescence & Phosphorescence::Persistent afterglow",
1
],
"color": "#777d2b",
"bgcolor": "#636917"
},
{
"id": 549,
"type": "PrimereCustomStyles",
"pos": [
4317.311750382618,
1360.0800425843127
],
"size": [
340,
414
],
"flags": {},
"order": 38,
"mode": 0,
"inputs": [
{
"name": "opt_pos_style",
"shape": 7,
"type": "STRING",
"link": 5738
},
{
"name": "opt_neg_style",
"shape": 7,
"type": "STRING",
"link": 5739
}
],
"outputs": [
{
"name": "STYLE+",
"type": "STRING",
"links": [
5740
]
},
{
"name": "STYLE-",
"type": "STRING",
"links": [
5741
]
}
],
"properties": {
"Node name for S&R": "PrimereCustomStyles"
},
"widgets_values": [
"human_trace_style.toml",
"Human Presence Without Humans::Recently occupied",
1,
"None",
1,
"None",
1.0029999999999997,
"None",
1,
"Temporal Trace::Aged trace",
1
],
"color": "#777d2b",
"bgcolor": "#636917"
},
{
"id": 548,
"type": "PrimereCustomStyles",
"pos": [
3956.4811553153468,
1359.5925719455224
],
"size": [
340,
414
],
"flags": {},
"order": 25,
"mode": 0,
"inputs": [
{
"name": "opt_pos_style",
"shape": 7,
"type": "STRING",
"link": null
},
{
"name": "opt_neg_style",
"shape": 7,
"type": "STRING",
"link": null
}
],
"outputs": [
{
"name": "STYLE+",
"type": "STRING",
"links": [
5742
]
},
{
"name": "STYLE-",
"type": "STRING",
"links": [
5743
]
}
],
"properties": {
"Node name for S&R": "PrimereCustomStyles"
},
"widgets_values": [
"emotions_style.toml",
"None",
1,
"Melancholic",
1,
"None",
1,
"Surprise::Astonishment",
1.0029999999999997,
"None",
1,
"Sadness::Depression",
1,
"None",
1
],
"color": "#777d2b",
"bgcolor": "#636917"
}
],
"links": [
@@ -11716,8 +11742,8 @@
"bounding": [
3943.895263671875,
1284.5947265625,
1083.4688720703125,
1134.0596923828125
1085.563245210088,
1143.1351559300515
],
"color": "#808040",
"flags": {}
@@ -11750,10 +11776,10 @@
"config": {},
"extra": {
"ds": {
"scale": 0.7972024500001696,
"scale": 0.9646149645002058,
"offset": [
-13.378027010046026,
179.71846983105775
-619.015884682149,
209.80806738458145
]
},
"frontendVersion": "1.45.19",