NIND Denoising

Previously, RapidRAW only had a traditional BM3D denoiser. It produced decent results, but it was never really competitive with the newer AI-based denoising methods found in other software.

So I'm excited to introduce a new denoising method based on nind-denoise. It can be applied directly to RAW images to naturally remove noise while preserving detail.

The results speak for themselves:

Original	BM3D	NIND

Due to dependency limitations, NIND denoise is not yet GPU accelerated, so it will take some time to process each image. In future versions, once things have matured, I plan to support batch denoising and move the processing over to the RapidRAW AI Connector, which supports GPU acceleration.
Credits for this denoising method go to @trougnouf's nind-denoise.

Image Analytics Panel

RapidRAW now features a switchable image analytics panel, giving you multiple ways to exactly visualize what's going on with your edit in real time.

Supported modes:

• Luma Waveform
• RGB Waveform
• RGB Parade
• Vectorscope
• RGB Histogram

Parametric Color & Luminance Masks

I've added two new mask types that let you quickly target specific color ranges or luminance zones in your image.

Single Point Subject Masks

This release ships with updated, more accurate, and higher resolution AI subject detection models based on SAM-2. On top of that, I've implemented single point masking - just click on your subject and it gets selected. No need to draw a bounding box around it first.

JPEG XL, WebP, and Additional Format Support

This update adds support for a bunch of new export formats, including JPEG XL, WebP, and CUBE LUTs. The LUT export is something I'm particularly happy about: You can screenshot a frame from your video editor, color grade it in RapidRAW, export the result as a LUT, and then apply that same color grade back in the video editor at full quality.

LRU Cache

RapidRAW now keeps recently viewed images in a cache, so switching between them is instant and seamless. This is especially useful when you're culling and need to quickly flip back and forth to check focus.

ROI Rendering

Previously, dragging sliders while zoomed in would lag and stutter because the entire image was being reprocessed on every change. Now RapidRAW only processes the region that's actually visible in the viewport, which gives a huge speedup when editing zoomed in.

• Improved tonemapper & shader, particularly how it behaves within masked regions
• Refined keybinds across the app for a smoother editing workflow
• Various UI polish and visual tweaks
• A ton of bug fixes under the hood - too many to list individually ;)

Rewritten Rendering Engine

I've rebuilt the viewport & rendering engine to offer two distinct ways to preview your images:

1. Dynamic Preview

This mode renders the image pixel-accurate to your specific display. Whenever you zoom in, the image dynamically updates to stay perfectly sharp. I’ve also added a resolution scaling option: this allows you to lower the render scale to improve performance on high-resolution (4K/5K) displays where you might not need the absolute highest pixel density at all times.
While building this, I managed to fix a long-standing WebKit (macOS) bug that was causing images to appear blurry while zooming.

2. Static Preview

This mode renders the image based on a fixed base resolution that you set. It won't dynamically recalculate every time you zoom (though it will still load a full-resolution image when you enter fullscreen mode). This option is highly recommended for lower-end devices where keeping the app snappy is the top priority.

Native Viewport Fullscreen

Previously, entering fullscreen mode in RapidRAW would open a separate, high-resolution image modal, which never quite felt native. Now, when you enter fullscreen, the side panels smoothly slide out of the way and the main viewport scales up seamlessly. Because it uses the actual viewport, all standard zoom, pan, and editing controls now work perfectly while in fullscreen.

Fluid Mask Creation & Real-Time Overlays

The semi-transparent red mask overlay is now calculated in true real-time. This allows you to make micro-adjustments to your masks without any stuttering or delayed preview updates.

I've also made creating new masks much more intuitive: instead of Linear and Radial masks automatically dropping into the center of your image, the tool now lets you click and drag to draw them exactly where you want them from the very first click.

True Pixel Peeping (No Interpolation)

For those who need to closely inspect their images, RapidRAW now disables smoothing/interpolation when you reach the maximum zoom level. This allows you to see the raw, hard-edged pixels to perfectly analyze noise and sharpness.

• Export Memory: RapidRAW now remembers your last-used export settings so you don't have to set them up every time.
• Auto-Crop on Lens Correction: Applying lens correction will now automatically crop the image slightly to hide the black borders and distorted edge artifacts.
• Real-Time Histogram: The histogram now updates instantly and smoothly as you adjust sliders.
• Batch Negative Conversion: You can now convert negative film scans in batches, speeding up film workflows.
• Various under-the-hood shader improvements for better performance.
• General UI cleanup, Lensfun bump and stability fixes.

Automatic Negative Conversion

I’ve added a new productivity tool to accurately invert film scans with a single click. This feature automatically analyzes the film base and inverts the colors to a positive image. This implementation wouldn't be possible without @marcinz606's NegPy.

Film Scan

Automatically Converted

HDR Merge

You can now merge multiple exposures into a single High Dynamic Range image directly within the app. This allows you to combine bracketed shots to recover highlights and shadows that a single RAW file couldn't capture.

Here is an exposure bracket of 3 images:

The result is a image file containing the full dynamic range of all 3 images, ready for editing:

Detail-Aware Tonal Adjustments

I have significantly improved how basic tonal adjustments (like Shadows/Highlights) interact with image textures.

Previously, these adjustments didn't distinguish well between large tonal areas and fine local contrast, which could lead to flat or "washed out" textures when recovering shadows. I’ve now implemented tonal detail masks into the pipeline. The engine now accurately targets luminance levels while protecting local contrast and fine details.

Original

v1.4.12 (Basic)

v1.5.0 (Detail-Aware)

Custom AI Tagging

You can now define your own list of keywords, and the AI will analyze your photos and apply the most relevant tags automatically. This is great for culling large shoots. For example, if you input "day, night, indoor," the AI will sort your library accordingly, allowing you to filter by specific scenes instantly.

XMP Metadata Read/Sync

The app now supports full XMP sidecar synchronization. This means you can read and write metadata (Ratings, Color Labels, and Tags) seamlessly, ensuring compatibility with other photo editors and asset managers. This behavior can be toggled in the settings.

Dropdown History

Right-clicking the Undo or Redo button now reveals a dropdown list of your edit history, making it easier to jump back to a specific point in your editing process without mashing Ctrl+Z.

New Composition Guide Overlays

I've added several new composition overlays to the crop tool. You can now cycle through various guides (Golden Ratio, Diagonals, etc.) to help nail the framing.

• Improved Image Canvas: Added single-click zoom, improved state memory (it remembers your previous zoom level), and implemented smoother exponential zooming.
• Performance: Faster, non-blocking fullscreen preview, improved image loading times and optimized live previews.
• Window State: The main window now remembers its size and position between sessions.
• Instant Metadata Loading: Metadata parsing is now significantly faster.
• Brush Tool: You can now draw straight mask lines by holding Shift + Click.
• Database: Enhanced Lensfun DB parsing and improved lens separation logic.
• Export: Added the ability to export masks as separate image files.
• LineaRAW: Exposed a setting for Linear RAW mode to fix pink casts and color artifacts.
• Bug Fixes: Fixed many, many, many bugs.

Creative Effects: Glow, Halation & Lens Flare

I have added three new adjustment tools to help level up your editing workflow. You can now use Glow to soften and bloom light sources, apply Halation to simulate the distinct red-orange bleeding found in analog film, or add Lens Flare to mimic light scattering through the glass elements.

Example of all 3 effects

Example of Lens Flare

Global Tooltips

RapidRAW now includes a native tooltip system. To help with accessibility and feature discovery, context menus now appear when hovering over buttons, explaining exactly what each control does.

Sharpness & Clarity Logic

I identified and fixed a rendering bug in the GPU processor where the "Structure" effect was incorrectly being applied to the Clarity and Sharpness sliders simultaneously. These controls are now decoupled and function independently as intended. If you have previously used the "Clarity" and "Sharpness" adjustments, your edits in these aspects may need some readjustment.

• Major UI polish: General cleanup of interface elements for a smoother & more polished experience.
• Rawler Sync: Added support for some more cameras & you're now able to load LinearRAW DNG images.

New RAW Preprocessing Algorithm

I've implemented a whole new raw preprocessing algorithm. It efficiently removes color noise and sharpens the image, offering a massive step up in quality in the whole pipeline compared to the previous version.

This update is also exciting for Fuji X-Trans users, as the new algorithm handles the sensor pattern much better, resulting in cleaner and sharper details.

Here is a comparison of 2 very noisy images alongside a Fuji X-Trans image at the end:

Before (Old Algo, Lots of Color Noise)	After (New Algo, Only Luma Noise)

Flat-Line Clipping in Curves

I've implemented flat-line clipping logic for the curves tool. You aren't limited to just lifting or pushing down curve points anymore; you can now drag the start and end points along the X-axis to clip the blacks or whites exactly how you need to.

• Improved Lens Detection: I improved the Lensfun parser to detect lenses more reliably and added support for more manufacturers.
• Smoother Image Navigation: Rapidly switching images (like holding down arrow keys) is now much more fluid. I implemented logic to cancel pending load requests, so the app doesn't get slowed down trying to open images you've already skipped past.
• Minor Changes: Added curves cross-channel copy/paste, productivity context menu is now also shown in the editor view, wgpu improvements, RAW tonemapping on non-edited images fixed & more...

Reworked Local Contrast Adjustments

The previous implementation of local contrast adjustments (e.g. sharpness, clarity & structure) had a nasty habit of creating "black holes" around high contrast edges. Not ideal when you're trying to make your photos look good...

I've completely reworked the shader with a different approach that pretty much obliterates these issues:

Before (0)	Before (+100 Sharpness)	After (+100 Sharpness)

New Fine Rotation Grid

When you're fine-rotating or transforming an image, the grid now transitions from the rule of thirds to a detailed fine grid. Because sometimes you need more precision than kinda straight.

Favourite Lens Saving & Improved Correction Accuracy

You can now save your frequently used lenses in the settings panel for quick access in the Lens Correction panel. No more scrolling through the entire database of lenses humanity has ever created.

On top of that, I've improved the accuracy of the lens correction itself based on some initial feedback and testing.

• Fixed quick eraser functionality
• Improved sepia theme contrast

Lens Correction

Proper lens correction has been one of the most requested features, and I’ve finally integrated the Lensfun library. This allows for dedicated corrections of optical flaws:

My knowledge of lens correction is limited, so some aspects / sliders may require further refinement in future updates. I would appreciate your feedback.

New Geometry Warping Tools

I’ve built a dedicated transform modal to handle image geometry. This is about fixing perspective and composition. When you’re warping or transforming an image, the UI can display dynamic lines to help you align horizons and vertical structures perfectly:

A More Powerful Masking System

The old masking panel was starting to feel a bit limited and not really intuitive, so I’ve completely refactored it.

• Sequential Masking: Masks (& submask components) are now applied one after another. This makes the math much more predictable when you're stacking multiple adjustments.
• Drag & Drop: You can now reorder your mask layers just by dragging them. I also added smooth animations for adding or removing components so the UI doesn't feel so "jumpy."
• Per-Mask Opacity: I added individual opacity and invert toggles for every mask component, giving you much finer control over local adjustments.

Export Presets

You can now save your export settings. I got tired of re-typing the same JPEG quality and dimensions every time, so I built a manager for it:

Generative AI Infrastructure

I moved away from the direct ComfyUI integration in favor of a dedicated AI middleware client, which makes the overall architecture more robust and flexible.

The new RapidRAW-AI-Connector sits between RapidRAW and your local ComfyUI server and is responsible for image caching, workflow injection, and AI coordination. High-resolution images are cached once; for subsequent edits, only the mask and prompt are sent, significantly improving performance during iterative generative edits.

This system is very much work in progress and not ready for production use yet.

By decoupling the AI integration from RapidRAW itself, development of the AI Connector can now happen independently and in parallel. Updates to the connector no longer require re-downloading RapidRAW, making experimentation and rapid iteration much easier while preserving full control over hardware, models, and custom ComfyUI workflows.

Better Performance for Large Libraries

If you have thousands of photos in a folder, you probably noticed the filmstrip could get sluggish which then slowed down the whole editor. It now only renders what is actually on your screen, which makes navigating huge catalogs significantly faster.

Centralized EXIF

I refactored the metadata handling to ensure greater consistency across various RAW formats, enhancing support through Rawler. As a result, RapidRAW can now read EXIF data from additional specialized RAW formats, including .RAF and .CR3.

• Library Preloading: The app now tries to stay one step ahead of you by preloading images in the library to reduce loading times.
• Improved Contrast: I tweaked the dark theme to make the UI elements pop a bit more.
• Consistent UI: I’ve standardized the slider styles across all panels and fixed many alignment issues or other inconsistencies.
• Updated Rawler: The raw loading library is again synced and updated, which adds support / improvements for some more cameras.
• A Lot More: As I've decided to rather use my time for coding than to write patch notes, this list is a bit shorter now ;)

Live Previews

Editing in RapidRAW is now more fluid and responsive than ever. Previously, adjustments were applied only when you stopped dragging a slider. Now, you can see your changes happen in near real-time as you drag, providing instant visual feedback.

This was made possible by a large rewrite of the GPU processing engine and improvements to the architecture in general. If you prefer the old behavior or are using a less powerful machine, you can disable live previews in the settings.

Before (Update After Dragging)	After (Live Previews Update)

Updated Collage Maker

The collage maker has been completely overhauled to give you more creative control. Key upgrades include:

• Drag & Drop: No more random mixing! Place your images exactly where you want them.
• Keep Aspect Ratio: A new option to prevent unwanted cropping of your images.
• Zoom & Pan: Finetune the composition within each cell.

LUT Quality Improvement

Previously, applying some LUTs could result in visible banding or posterization artifacts. This has been fixed by implementing tetrahedral interpolation in the shader, which ensures smooth, high quality color transitions even with lower-resolution LUT files.

Before (Banding Artifacts)	After (Smooth Interpolation)

• Separate preview worker, optional high-quality live previews & mask/ai patch caching.
• Enhanced EXIF UI including new infos when hovering the image name in the editor toolbar.
• 'Prefer RAW' filter option added to the library.
• Support for uppercase file extensions.
• Thumbnail cache is now flushed on folder switch.
• AI masking stability improvements under load.
• Rawler updated.

Merry Christmas everyone! 🎄 I wanted to get the new denoiser and other improvements out to you before the holidays start :)

Image Denoising

I previously removed the noise reduction controls because the old algorithm killed performance and didn't produce good results. I've now implemented BM3D (Block-matching and 3D filtering) as a dedicated image productivity tool. It is much more effective at removing noise while preserving actual image detail.

White Balance Picker

Added a pipette tool to the White Balance section. You can now pick a neutral reference point in the image to automatically calculate and set the correct temperature and tint.

Reworked Metadata Handling

Hovering over the file name in the toolbar now expands a quick-info widget displaying key shooting data: Aperture, Shutter speed, ISO, and focal length.

I also completely rewrote the backend for metadata saving. It now reliably writes EXIF data to the file during export (if enabled), fixing previous bugs where metadata was often lost.

Reworked Grain

I completely reworked the film grain engine. The previous version suffered from RGB color noise and a visible repeating grid pattern. I fixed the algorithm to generate natural, truly random monochromatic noise without artifacts.

Delete Workflow Improvements

Deleting an image no longer forces the library list to reload. This makes culling and cleaning up your folders significantly smoother.

• Improved mask eraser tool behavior
• ORT v2 migration
• Exposure slider added to masking tools
• Batch export stability improvements & RAM optimization
• New export option added directly to the context menu
• Updated dark mode wallpaper

Recursive Library View

You can now display an image grid that includes files from all subfolders within a selected directory. This makes it much easier to keep an eye on your entire image library at once. Performance is optimized for about 2-3k images.

Reworked HSL Luminance Controls

I have reworked the luminance logic in the HSL panel. Previously, it tended to create strange, blocky artifacts on compressed images (and even some RAWs). The new implementation results in much smoother transitions.

Before (Harsh falloffs & artifacts)	After (Natural falloffs & clean)

Mask Stacking

Previously, adding a new mask on top of an existing one would visually reset the changes made by the previous mask. This is now fixed: adding a new empty mask on top of another simply adds to the stack without altering the base state.

Comparison using a linear gradient (+5 exposure) followed by an empty radial mask:

Before (Visual reset bug)	After (Correct stacking)

Vibrancy Improvement

The vibrancy slider algorithm has been tweaked to better protect skin tones from oversaturation.

• QOI Support: Added support for the QOI file format.
• Camera Support: Updated the raw loader rawler library to support a wider range of cameras.
• DNG Flexibility: Improved handling of DNG files.
• Bug Fixes: Fixed several bugs within the masking system.