iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
Affinity Photo does not currently document a one-click AI layer splitter. Its machine-learning tools create pixel selections for a named subject, object, or scene region. You then turn that selection into a layer mask, copy the isolated pixels to another layer, and refine the edges. Manual Selection Brush work, Quick Mask, edge refinement, channel masks, and saved selections remain essential when the image contains hair, holes, fine structures, overlapping objects, or features the model does not recognize.
What Affinity Photo’s AI actually does
Affinity Photo V2.6 introduced the Object Selection Tool and Select Subject as its first machine-learning features. The model is downloaded to the device and runs on-device; Affinity’s support information says user data does not leave the device for model training. These tools analyze a pixel layer and return a pixel selection. They do not inspect a flattened photograph and automatically build a stack of semantically named, independently editable layers.
That distinction matters. A selection is a temporary boundary describing which pixels are active. A mask uses that boundary to reveal or hide pixels on a layer while preserving the original image underneath. A true layer-splitting command would have to identify every relevant object, create separate layers, name them, and resolve overlaps and transparency. The documented AI workflow stops at the selection stage, so you control how the result becomes editable artwork.
Free tools Windows power users keep installed
One-click scans. No signup required.
AI selection coverage and limits
Detected Objects
Affinity Help describes Detected Objects as recognizing individual instances across 80 everyday object categories. You can request an object by name instead of painting around it or sampling a color. If several instances are present, the tool can target an individual detected instance rather than treating the entire photograph as one undifferentiated subject.
#1 Best Overall
- Read Before You Buy — No Video Output: These adapters support charging and USB 2.0 data transfer, but cannot transmit video signals. Except for standard USB webcams (which use USB data only), they are not compatible with HDMI/DisplayPort cables, video-capable USB-C hubs, or docking stations with video output.
- Convert USB-A Ports to USB-C: Designed to connect USB-C earphones, cables, flash drives, card readers, and other USB-C accessories to standard USB-A ports. Plug-and-play with no drivers or software required.
- Aluminum Alloy Housing: Built with a sturdy aluminum alloy shell that aids in heat dissipation and protects against daily wear and scratches. Designed to maintain a stable and secure connection.
- Compact & Travel-Friendly: The ultra-compact design allows the adapter to stay plugged into your device without blocking adjacent ports or adding bulk, reducing wear and tear on your original USB ports.
- 12-Month Warranty: Backed by a 12-month manufacturer warranty for peace of mind. Designed to meet strict quality control standards for reliable everyday performance.
Detected Regions
Detected Regions covers eight scene categories: Sky, Foliage, Grass, Water, Mountains, Ground, Buildings, and Roads. Region detection is useful for broad tonal or compositional edits, such as masking the sky for a replacement or adjusting water without affecting the shoreline. It still returns a selection, not a region layer with independent geometry.
Recognition is not universal
Anything outside the supported object categories or region classes may be missed, merged with a neighboring item, or selected only partially. A model can also make a plausible but unsuitable boundary where objects overlap, where translucent material crosses a background, or where detail is smaller than the image resolution. Manual techniques are not limited to a fixed category list: you can target an arbitrary logo, cable, architectural detail, shadow, color range, or negative space.
Plan and hardware availability
Affinity currently marks Detected Objects and Detected Regions as premium features in its Help documentation. Confirm that your edition and account include them before planning a production workflow. Machine-learning support also depends on the device:
Recommended Free Tools
| Platform | Published requirement for ML features | Unsupported hardware called out in the requirements |
|---|---|---|
| Mac | Apple silicon (M-series) Mac running macOS Ventura 13 or later | Intel-based Macs |
| iPad | Apple silicon (M-series) iPad running iPadOS 17 or later | A-series iPads |
| Windows | Windows 10 or 11 on x64 or ARM64 | Devices that do not meet the supported Windows architecture requirements |
Requirements and plan entitlements can change with a product update, so check the current Affinity requirements and Help pages for your installed version before troubleshooting a missing tool.
Can you separate a subject into its own layer?
Yes, but you create the layer from the AI selection; Affinity does not automatically decompose the whole image for you. The safest result is a non-destructive masked layer. The original pixels remain available if you later discover that the selection clipped a finger, wheel spoke, strand of hair, or hole in the subject.
Rank #2
- 5-in-1 USB-C Hub: Experience comprehensive connectivity featuring a Power Delivery input, two USB-A 2.0 ports, a USB-A 3.0 port, and an HDMI port. (Note: The USB-C power delivery input port is only for connecting an external wall charger to power your laptop and cannot power peripheral devices.)
- 90W Pass-Through Charging: Achieve optimal charging with 90W pass-through power to your laptop, supported by a total input of 100W, with the hub reserving 10W for operational efficiency. (Note: Wall charger not included.)
- Quick Data Transfers: Accelerate your productivity with rapid data transfers using a high-speed 5Gbps USB 3.0 port and two 480Mbps USB 2.0 ports.
- 4K HDMI Display: Enhance your visual experience with a hub capable of delivering 4K resolution at 30Hz in both mirror and extend modes. Please note that this hub is compatible with MacBook (macOS 12 and newer), Windows 10 and 11, ChromeOS, and laptops equipped with DP Alt Mode and Power Delivery. Note: This device is not compatible with Linux.
- What You Get: Anker USB-C Hub (5-in-1, 4K HDMI), welcome guide, 18-month warranty, and our friendly customer service.
AI-to-mask workflow
- Open a pixel layer and make a backup. Duplicate the source layer in the Layers panel or keep the original hidden and untouched. This gives you a clean recovery point if later refinement goes wrong.
- Run Select Subject or Object Selection Tool. For a main foreground subject, use Select Subject. For a recognized individual item, use Object Selection Tool and choose the detected object. For a broad scene component, choose the appropriate Detected Region.
- Inspect the marching-ants boundary at high zoom. Look around hair, holes, thin limbs, bicycle spokes, overlapping objects, and contact shadows. Do not assume a named selection is production-ready.
- Convert the selection to a mask. With the target pixel layer active, add a mask layer from the Layers panel. The active selection becomes the mask’s revealed area; the hidden pixels stay on the layer for later correction.
- Refine the mask, not the source pixels. Paint white on the mask to reveal omitted detail and black to conceal spill. Use a soft, low-opacity brush where the edge should transition gradually, and a hard brush for crisp manufactured edges.
- Separate only when needed. If you need a movable isolated subject, duplicate the masked layer and keep the mask on the duplicate. You can then transform, recolor, or retouch it independently while preserving the original layer below.
- Save a reusable selection when the boundary has value beyond this edit. Store it as a spare alpha channel or export it as an external selection file. Saved selections can be reloaded into pixel selections, masks, adjustments, or live filters.
This procedure gives you the practical equivalent of a subject layer without pretending that the AI has created a complete semantic layer hierarchy.
Manual selection and masking methods
Selection Brush Tool
The Selection Brush Tool is the general-purpose manual alternative. Paint over the area to add it to the selection, switch to Subtract when the brush spills outside, and use edge snapping when the boundary has a strong contrast. The All Layers option lets you sample visible content across layers instead of restricting the calculation to the active layer. Refine controls help smooth, feather, or contract a boundary before you commit it to a mask.
The Tool Desk
Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Manual painting takes more interaction than asking AI for a named subject, but every stroke is intentional. That control is valuable for irregular products, small foreground details, repeated patterns, and objects that are not represented by the model’s categories.
Quick Mask and refinement
Quick Mask provides a colored overlay so you can paint the selection visually rather than relying only on marching ants. It is particularly useful for checking holes and inclusions: the opening in a handle, the gap between legs, or the space between branches should remain unselected when the subject is isolated. The Affinity Photo workbook that documents Quick Mask also covers selection-refinement techniques, making it a useful hands-on reference for learning the workflow.
Channel-based masks
Affinity documents channel masking as an alternative when Flood Select or Selection Brush would be slow or inaccurate. Choose the channel with the strongest contrast between the target and its background, load that channel as a pixel selection, and add a mask layer. You can then apply adjustments or filters non-destructively through the mask.
Rank #3
- Sleek 7-in-1 USB-C Hub: Features an HDMI port, two USB-A 3.0 ports, and a USB-C data port, each providing 5Gbps transfer speeds. It also includes a USB-C PD input port for charging up to 100W and dual SD and TF card slots, all in a compact design.
- Flawless 4K@60Hz Video with HDMI: Delivers exceptional clarity and smoothness with its 4K@60Hz HDMI port, making it ideal for high-definition presentations and entertainment. (Note: Only the HDMI port supports video projection; the USB-C port is for data transfer only.)
- Double Up on Efficiency: The two USB-A 3.0 ports and a USB-C port support a fast 5Gbps data rate, significantly boosting your transfer speeds and improving productivity.
- Fast and Reliable 85W Charging: Offers high-capacity, speedy charging for laptops up to 85W, so you spend less time tethered to an outlet and more time being productive.
- What You Get: Anker USB-C Hub (7-in-1), welcome guide, 18-month warranty, and our friendly customer service.
Channel masking often outperforms color sampling on hair, foliage, smoke, glass, and other high-frequency edges when one color channel separates the subject more clearly than the composite image. It is not an automatic solution: you may still need to invert, paint, or combine channels to address uneven lighting and background contamination.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Saved selections and spare channels
A selection that took time to perfect should not be treated as disposable. Save it as a spare alpha channel or an external selection file. Reloading the saved boundary lets you create a new mask, adjustment, or live filter without repainting. This is the repeatability advantage of a disciplined manual workflow, especially when a client requests several color treatments or revisions months later.
AI versus manual methods: practical comparison
| Decision factor | AI selection | Manual selection and masking |
|---|---|---|
| Initial isolation speed | Can produce a named subject, object, or region selection quickly; no controlled time-saving benchmark is published. | Requires painting, sampling, or tracing, so setup usually involves more interaction. |
| Edge and exception control | Provides a useful first boundary but can need correction around hair, holes, translucency, and overlaps. | Selection Brush, Refine, Quick Mask, and channel masks provide direct control over difficult boundaries. |
| Editability | Ends in the same non-destructive mask workflow as a manual selection. | Also supports masks that hide or reveal pixels without erasing the source. |
| Repeatability | The generated selection can be saved after correction, but the detection itself may vary with image content. | Spare channels and external selection files preserve a carefully edited boundary for reuse. |
| Coverage | Limited to recognized objects and the eight documented region classes. | Can target any visible feature, regardless of category. |
| Availability | Requires supported ML hardware and may require a premium plan for Detected Objects and Detected Regions. | Uses the standard selection and masking tools and does not depend on ML detection. |
The most defensible conclusion is not that one method is universally better. AI is an efficient starting point when the image contains a recognizable subject or region. Manual work is the finishing and fallback path when precision, arbitrary coverage, or repeatability matters.
Which workflow should you choose?
Typical portrait or product on a simple background
Start with Select Subject, convert the result to a mask, and inspect the outline at 100–200% zoom. Use the Selection Brush or mask painting for missed edges. This avoids tracing a boundary that the model can identify cleanly while retaining manual control over the final pixels.
Hair, fur, foliage, smoke, or translucent edges
Use AI only as a rough starting point, then switch to Refine, Quick Mask, and mask painting. Check both the outer contour and the background showing through the subject. If the background and subject separate strongly in one channel, build a channel-based mask instead of repeatedly brushing a low-contrast edge.
Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Fix the driver behind crashes, sound loss and screen glitches3Clear out junk files and repair common Windows errorsRank #4
- Dual Converters, Infinite Potential：Includes 2× USB C male to USB A female adapters and 2× USB A male to USB C female adapters. Perfect for a wide range of uses—tablets with Bluetooth keyboards, expand USB ports on macbook, and more. Two different converters for all your daily needs
- Next-Level 10Gbps & 3A Charging: No more slow 480Mbps, this usb to usb c adapter has a transfer speed of up to 10Gbps, allowing you to do more transferring in less time. This usb adapter fits both USB A and USB C charger, supporting up to 3A fast charging
- Upgraded Exquisite Craftsmanship: With an aluminum alloy housing and metal connector, the usbc to usb adapter is extremely durable and sturdy. Rigorously tested to withstand more than 10,000 times of plugging and unplugging, ensuring long-lasting performance
- Broad Compatible: The usb c to usb adapter widely supports all USB C/ USB A devices like laptops, tablets, cellphones, car chargers, and phone chargers. Such as compatible with MacBook Pro/Air 2023/2022, Thunderbolt 4/3 Devices,Apple MagSafe Watch 9/8/7/SE/Ultra, iPad Pro 2022/2021, Samsung Galaxy S23/S20/S10, and iPhone 17/16/15 Pro. Plug and play
- Please Note: To reach 10Gbps speed, keep the cable under 3.3 ft. For USB A Male to USB C adapters, try flipping the USB C connector. USB C Male to USB A adapters support bidirectional 10Gbps transfer within 3.3 ft
Objects that overlap or touch
Run Object Selection Tool on each recognized instance separately when possible. Keep each result on its own masked duplicate so you can correct the overlap independently. If the detector merges the items, create a broad initial mask and divide the boundary manually with Selection Brush or Quick Mask.
Sky, water, roads, or other broad scene regions
Detected Regions can give you a quick starting selection for the documented classes. Examine the horizon, treelines, shorelines, and structures crossing the region. Manual painting or a channel mask is usually required where the region has fine interruptions.
Unusual subject or production template
Skip AI when the target is not a supported category, when the selection must be pixel-identical across many versions, or when the image contains repeated technical details. Build the mask manually, save it as a spare channel or external selection, and reuse that asset for subsequent adjustments.
How to refine an AI selection without damaging the image
- View the mask itself. Temporarily display the mask or use Quick Mask so gray, semi-transparent, and missing areas are visible instead of hidden by the composite image.
- Correct large errors first. Use a broad brush in Add mode to restore a missing body area or Subtract mode to remove a large background patch. Small brushwork is inefficient until the overall silhouette is right.
- Work around exclusions. Paint inside holes and between overlapping objects in Subtract mode. Check handles, straps, spokes, branches, and the spaces between fingers.
- Choose edge hardness deliberately. A hard brush preserves a manufactured edge; a softer brush suits hair, fabric, and atmospheric transitions. Avoid feathering every edge, because it can create a visible halo.
- Inspect against contrasting backgrounds. Place a temporary dark fill behind a light subject and a light fill behind a dark subject. Halos and background color spill that are invisible on the original often become obvious.
- Keep the source layer intact. Edit the mask or a masked duplicate rather than erasing pixels. If the edge needs another pass, you can return to the original data.
- Save the finished selection. Store a spare alpha channel or external selection file before creating multiple adjustments. This turns a one-off correction into a reusable production asset.
Common failure modes and fixes
The AI command is missing
Check that you are running a version that includes the V2.6 machine-learning features, that your operating system and hardware meet the published requirements, and that your edition includes the premium detected-selection features. Intel Macs and A-series iPads are specifically excluded from the ML requirements. On Windows, verify that the installation is running on supported x64 or ARM64 hardware.
The result selects the whole scene instead of the intended object
Use Object Selection Tool for an individual detected instance rather than a broad subject or region command. If the item is not recognized, switch to Selection Brush, Quick Mask, or a channel mask; manual selection is not restricted to the model’s category list.
Best Value
- 5-in-1 Connectivity: Equipped with a 4K HDMI port, a 5 Gbps USB-C data port, two 5 Gbps USB-A ports, and a USB C 100W PD-IN port. Note: The USB C 100W PD-IN port supports only charging and does not support data transfer devices such as headphones or speakers.
- Powerful Pass-Through Charging: Supports up to 85W pass-through charging so you can power up your laptop while you use the hub. Note: Pass-through charging requires a charger (not included). Note: To achieve full power for iPad, we recommend using a 45W wall charger.
- Transfer Files in Seconds: Move files to and from your laptop at speeds of up to 5 Gbps via the USB-C and USB-A data ports. Note: The USB C 5Gbps Data port does not support video output.
- HD Display: Connect to the HDMI port to stream or mirror content to an external monitor in resolutions of up to 4K@30Hz. Note: The USB-C ports do not support video output.
- What You Get: Anker 332 USB-C Hub (5-in-1), welcome guide, our worry-free 18-month warranty, and friendly customer service.
Fine detail disappears
Inspect the mask at high zoom, then add the missing strands or structures with a small brush in Add mode. For hair, foliage, or smoke, test a contrasting channel and use channel masking to build a more detailed boundary. Keep semitransparent transitions rather than forcing them into a hard black-and-white edge.
Background color remains around the subject
Display the mask against a contrasting temporary background and subtract the contaminated fringe. Reduce brush opacity for gradual cleanup, and avoid shrinking the entire mask aggressively because that can cut off legitimate fine detail.
A hole inside the subject is filled
Use Quick Mask or mask painting in Subtract mode inside the opening. Check every enclosed area before transforming or exporting the isolated layer.
You need the same boundary for several edits
Save the corrected selection as a spare alpha channel or external selection file. Reload it whenever you create another mask, adjustment, or live filter instead of rebuilding the boundary.
What “non-destructive” means here
A mask controls visibility; it does not delete the hidden pixels. You can disable, repaint, replace, or remove the mask and recover the underlying layer. This is true whether the mask began with Select Subject, Object Selection Tool, Selection Brush, Quick Mask, or channel information. Permanent pixel deletion is a separate decision and is unnecessary for the ordinary subject-isolation workflow.
Bottom line for choosing AI or manual masking
Affinity Photo’s AI features are best understood as selection accelerators, not an automatic layer-splitting system. Use them when a recognizable subject, object, or documented scene region gives you a sensible first boundary. Convert that selection to a mask and refine it. Choose manual Selection Brush work, Quick Mask, channel masking, or a saved spare channel when the edge is complex, the target is outside the model’s categories, or the result must be repeatable. Both approaches can produce editable, non-destructive layers; the difference is how much of the boundary the software can identify before you take over.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
