AI Editor

The AI Editor transforms natural-language text prompts into images and edits. Simply describe what you want to create or change, and the AI Editor generates visual results based on your description.

What you can do:

Generate new images from text descriptions with multiple model options
Edit existing images using natural language or uploaded references
Control generation settings (aspect ratio, number of variations)
Upload images for reference or editing, including camera capture
Remove or replace backgrounds
Add realistic shadows and lighting
Upscale images and remove watermarks

The AI Editor combines the simplicity of text-based commands with precision controls for targeting specific regions and refining results.

Quick Start

Generate an Image From Scratch

Navigate to Studio in your PixelBin Dashboard
Select your preferred model and aspect ratio
Choose the number of images to generate
Enter a text prompt describing your desired image
Example: "a modern office workspace with natural lighting"
Optionally upload a reference image to guide the style
Click Generate
Review variations and refine your prompt or settings as needed

Edit an Existing Image

Navigate to Studio in your PixelBin Dashboard
Upload an image from your device or capture with camera
Select your preferred model and generation settings
Enter a prompt describing the change you want
Example: "replace the background with a sunset beach scene"
Click Generate to apply the edit
Iterate by adjusting your prompt, settings, or using selection tools for precision

tip

Start with simple, clear descriptions. You can always refine and add detail in subsequent generations.

How It Works

The AI Editor uses machine learning models to interpret your text prompts and generate corresponding visual output. Each generation is probabilistic, meaning:

The same prompt may produce different variations
Small changes in wording can affect the result
Results improve through iteration and prompt refinement

Think of prompts as creative instructions rather than exact commands. The AI Editor excels at rapid experimentation and creative exploration.

Use Cases

The AI Editor is designed for creative workflows that benefit from speed, flexibility, and natural-language control.

Ideal for:

Rapid prototyping and concept exploration: Generate multiple variations quickly to explore visual directions
Background manipulation: Remove, replace, or enhance backgrounds without manual masking
Non-destructive editing: Make changes while preserving the ability to iterate or revert
Creative experimentation: Test ideas that would be time-consuming with traditional tools
Content adaptation: Modify existing images for different contexts or platforms

Best results when:

The desired outcome can be clearly described in text
Some variation between generations is acceptable
Iterative refinement is part of the workflow
Speed and exploration matter more than pixel-perfect precision

Understanding AI-Generated Results

The AI Editor uses machine learning to transform your text descriptions into images. It interprets your prompts and generates images by recognizing patterns learned from large pool of training data, giving you creative flexibility and rapid iteration capabilities.

Creative exploration: Generate multiple unique variations from the same prompt to explore different visual directions
Flexible interpretation: The AI adapts to your wording, allowing you to guide results through natural language
Iterative refinement: Each generation is an opportunity to adjust and improve, making the creative process dynamic

What to expect:

Unique variations: Each generation produces fresh results, even with identical prompts
Prompt sensitivity: Adjusting your wording gives you control over the output direction
Evolving results: Refine prompts based on what works to achieve your desired outcome

Writing Effective Prompts

Well-crafted prompts lead to better results. This section provides practical guidance on structuring prompts, common pitfalls to avoid, and examples to inspire your own creations.

Prompt Structure

Effective prompts typically include these components:

Subject: What you want to create or modify (person, object, scene)
Context: Setting, environment, or background details
Style: Visual aesthetic, artistic style, or mood
Quality modifiers: Details about lighting, composition, or technical aspects

Simple formula:

[Subject] + [Context] + [Style] + [Quality modifiers]

Example:

"a vintage camera" + "on a wooden desk with morning light" + "minimalist product photography" + "soft shadows, high detail"

Do's and Don'ts

Do:

Be specific: "a golden retriever puppy" is better than "a dog"
Use descriptive language: Include colors, textures, materials, and atmosphere
Specify style when needed: "photorealistic," "watercolor painting," "3D render"
Include important details: Lighting, perspective, mood, time of day
Start simple, then refine: Add details gradually based on results
Use clear, natural language: Write as you would describe the image to someone

Don't:

Be vague: "nice background" doesn't provide enough direction
Overload with contradictions: "bright and dark" or "minimalist but detailed" confuse the model
Use ambiguous references: "like that famous painting" won't work without specifics
Expect literal interpretation: The AI predicts patterns, not exact specifications
Stuff keywords: Natural sentences work better than comma-separated lists
Assume context: The AI doesn't know your previous attempts or intentions

Prompt Examples

For Image Generation

Product Photography:

a sleek black smartphone on a marble surface, studio lighting, professional product shot, sharp focus

handmade ceramic coffee mug with blue glaze, on a rustic wooden table, natural window light, shallow depth of field

luxury watch with leather strap, dramatic side lighting, dark background, commercial photography style

Creative & Artistic:

a futuristic cityscape at sunset, neon lights reflecting on wet streets, cyberpunk aesthetic, cinematic composition

abstract geometric pattern in teal and gold, art deco style, symmetrical design, high contrast

watercolor painting of a mountain landscape, soft pastel colors, misty atmosphere, loose brushstrokes

Realistic Scenes:

a cozy reading nook by a window, warm afternoon sunlight, bookshelves in background, comfortable armchair

modern minimalist kitchen, white marble countertops, natural wood accents, bright and airy

a peaceful forest path in autumn, golden leaves on the ground, soft diffused light through trees

For Image Editing

Background Replacement:

replace the background with a professional office environment, soft focus

change background to a tropical beach at sunset, warm lighting

replace with a solid gradient background, blue to purple

Object Modification:

change the car color to metallic red, keep everything else the same

remove the person on the left side of the image

add realistic shadows beneath the product to ground it on the surface

Lighting & Atmosphere:

add warm golden hour lighting to the scene

make the image darker and moodier, increase contrast

add soft rim lighting from the right side

Style Transfer:

convert to black and white with high contrast, film noir style

make it look like a vintage photograph from the 1970s, faded colors

apply a painterly effect, impressionist style

Refining Your Prompts

If results aren't what you expected, try these adjustments:

Issue	Solution
Too generic or vague	Add specific details about colors, materials, lighting, or style
Wrong style or mood	Explicitly state the aesthetic: "photorealistic," "cartoon," "cinematic"
Missing important elements	List key components you want included in the scene
Unwanted elements appear	Specify what should be in the image; try regenerating with clearer focus
Lighting is off	Add lighting descriptors: "soft natural light," "dramatic shadows," "bright and even"
Composition issues	Include framing terms: "close-up," "wide angle," "centered composition"

Quick Reference: Prompt Patterns

Pattern	Use Case	Example
[Subject] in [Setting]	Simple scenes	"a red bicycle in a park"
[Style] of [Subject]	Artistic renders	"watercolor painting of a sunset"
[Subject] with [Details]	Detailed generation	"a laptop with glowing screen, on a desk with plants"
Replace [X] with [Y]	Background swaps	"replace the sky with a starry night"
Add [Element] to [Location]	Adding elements	"add shadows under the product"
Make it [Adjective]	Style adjustments	"make it brighter and more vibrant"
[Subject], [Style], [Mood]	Complete description	"forest path, photorealistic, peaceful morning atmosphere"

tip

Save prompts that work well for you. Build a personal library of effective prompt patterns for different use cases.

Accessing the AI Editor

Navigate to Studio in your PixelBin Dashboard to access the AI Editor. From here, you can:

Enter a text prompt to generate an image or video
Click Generate to process your prompt
Continue editing in the AI Editor interface

Generation Settings

The AI Editor prompt window provides several controls to customize your generation:

Model Selection

Choose from available AI models to generate images. Different models offer varying characteristics in terms of output style, quality, and credit consumption.

Available models:

Nano Banana: Standard quality, 3 credits per generation
Seedream 4.5: Alternative standard quality, 3 credits per generation
Nano Banana Pro: Enhanced quality, 11 credits per generation

Selecting Auto allows the system to automatically choose a model based on the prompt and configuration. This option is useful when users are unsure which model best fits their use case.

Model selection affects the visual characteristics and detail level of generated images.

Aspect Ratio

Control the dimensions of generated images by selecting from preset aspect ratios optimized for different platforms and use cases:

1:1 (Square): Perfect for Instagram posts, profile pictures, and general social media content that works across platforms
16:9 (Widescreen): Ideal for YouTube thumbnails, website banners, presentations, and desktop wallpapers
9:16 (Social Story): Optimized for Instagram Stories, TikTok, Reels, and other vertical video platform content
4:5 (Social Post): Best for Instagram feed posts and Pinterest pins, maximizing vertical space in mobile feeds

Choose the aspect ratio that matches your intended platform or output requirements. Each ratio is designed to fit specific display contexts without cropping.

Generations

Specify how many image variations to generate per prompt. Generating multiple variations allows you to:

Compare different interpretations of the same prompt
Select the best result from multiple options
Explore creative variations quickly

Each generated image consumes credits according to the selected model.

tip

Generate multiple variations (2-4 images) when exploring new concepts. This gives you options to compare and helps identify which prompt elements work best.

Image Upload Options

The prompt window supports multiple ways to provide images as input for generation and editing.

Upload Files

Upload files directly from your device to use as input for generation and editing. The AI Editor supports both images and documents.

Supported file types:

PDF files
Extract content for text-based generation or analysis
Use document content as context for image generation

How to use uploaded files:

When uploading documents like PDFs, the AI Editor can extract and process the content to inform generation. This lets you to create images based on document content or use textual information from documents as context for your prompts.

Open Camera

Access your device camera directly from the prompt window to capture and upload images in real-time. This is useful for:

Quick photo-based editing workflows
Capturing reference images on the spot
Mobile-first creative processes

The captured image can be used immediately as a reference or for editing.

Upload Image

When you upload an image as a reference:

The AI uses it to guide the style, composition, or visual characteristics
Your text prompt still controls the primary content and subject
Reference images influence but don't guarantee exact matches

Example workflow:

Upload a reference image with a specific aesthetic
Enter a prompt describing your desired subject
Generate to see results influenced by both prompt and reference

note

Reference images guide the overall aesthetic and style. Combine them with clear text prompts for best results.

Editing Uploaded Images

When you upload an image for editing:

Enter a prompt describing the changes you want to apply
The AI modifies the uploaded image based on your instructions
Use selection tools (Smart Select, Brush Select) for targeted edits

This workflow is ideal for iterative refinement of existing images.

Editing Capabilities

Text-to-Image Generation

Create new images from scratch using text descriptions, with optional reference images to guide the style.

How it works:

Enter a text prompt defining the subject, environment, and style
Select model, aspect ratio, and number of images to generate (max 4 images from a single prompt)
Optionally upload a reference image to influence the aesthetic
Generate to create image variations

Example prompts:

a minimalist product photo of a coffee mug on a wooden table

an abstract geometric pattern in blue and gold

a futuristic cityscape at night with neon lights

note

Each generation is unique. Running the same prompt multiple times will produce variations. Generate multiple images per prompt to compare options.

Prompt-Based Image Editing

Modify existing images using text descriptions. Upload an image from your device or capture with camera, describe your desired changes, and the AI Editor applies them while preserving the original context.

How it works:

Upload your image or capture with camera
Select model and generation settings
Enter a prompt describing the edit
Optionally use reference images to guide the style
Generate to see the modified result

Common transformation types:

Structural: Remove or replace backgrounds, objects, or elements
Environmental: Change surroundings, add contextual elements, or modify settings
Lighting & Shadows: Generate or adjust shadows, modify lighting conditions
Stylistic: Alter colors, textures, or overall aesthetic

Example edits:

"change the background to a mountain landscape"
"add warm sunset lighting to the scene"
"remove the object on the left side"

tip

For precise edits to specific areas, use Smart Select or Brush Select to target regions before applying your prompt. Generate multiple variations to compare different interpretations.

Iterative Editing

Build complex edits by applying multiple transformations in sequence. Each generation uses the previous output as its starting point, allowing you to layer changes and refine results progressively.

Tips for iterative editing:

Make one type of change at a time for more predictable results
Save intermediate versions you're happy with
If quality degrades after several iterations, restart from an earlier version
Combine prompt edits with selection tools for precision control

Precision Tools

Beyond text prompts, the AI Editor includes specialized tools for targeted control. Use these to define where edits should apply or to perform specific image operations.

These tools work independently or alongside text prompts for enhanced precision.

Smart Select

Automatically identifies and selects the main subject or object in your image.

Use this selection to:

Constrain edits to specific regions
Prevent unintended changes to surrounding areas
Focus transformations on the main subject

Best for: Images with clear subject-background separation

Brush Select

Manually select specific areas by drawing over them with an adjustable brush.

When to use:

Automatic selection misses your target area
You need fine-grained control over the selection
You want to select multiple objects or complex regions

Combine with prompt-based edits for precise, targeted transformations.

Upscale Image

Increase image resolution by 2x, 4x, or 8x.

Available scaling factors:

2x: Moderate resolution increase, minimal artifact risk
4x: Significant resolution boost
8x: Maximum upscaling, may amplify existing artifacts

tip

Higher scaling factors may enhance imperfections. Start with 2x and increase if needed.

Remove Watermark

Remove visible watermarks using automatic detection or manual selection.

Auto Mode

AI-powered detection automatically finds and removes watermarks.

Options:

Remove text: Removes text-based watermarks
danger
This will remove ALL text in the image, not just watermarks.
Remove logo: Removes graphic watermarks while preserving other elements

Manual Mode

Draw over watermarks to remove them with adjustable brush size. Supports the same text and logo removal options as Auto mode.

When to use manual: Auto mode misses parts of the watermark or you need selective control.

Remove Background

Isolate your subject by removing the background entirely.

Object type selection:

Choose the category that best matches your subject for optimal results:

General: Default option for most images
E-commerce: Product photography
Car: Vehicle images
Human: People and portraits
Object: Specific items or objects

Selecting the correct type improves edge detection and subject separation quality.

Adjust

Fine-tune visual properties with standard image adjustments:

Brightness: Lighten or darken the image
Saturation: Increase or decrease color intensity
Contrast: Adjust the difference between light and dark areas
Warmth: Shift color temperature (cooler/warmer tones)

Adjustments apply to the entire image and can be combined with other edits.

Convert to Video

Transition your current image into the Video Generator to create motion-based content.

Your image becomes the starting frame for video generation, allowing you to add movement and animation using the existing visual as a foundation.

danger

This exits the AI Editor and opens the Video Generator as a separate workflow.