Photo Cutout: Extract Subjects Like a Pro

A clean photo cutout separates the subject from everything else in the frame. Product listings need white backgrounds. Designers need isolated objects for composites. Profile pictures need that distracting living room removed. Event marketers need speakers extracted for promotional graphics. The task is always the same: draw a precise boundary around the subject and discard the rest.

The challenge is also always the same: edges. Hair, fur, semi-transparent fabric, and fine details along the boundary are where every cutout tool either shines or falls apart. This guide covers the best tools for the job, when each one makes sense, and how to handle the difficult edge cases that trip people up.

Photo Cutout Tool Comparison

Before diving into workflows, here is how the major tools stack up for subject extraction:

Quick recommendation: For one-off cutouts with tricky edges, Photoshop is unmatched. For batch automation, Remove.bg's API or rembg are the fastest paths. For a free, private option that runs entirely in your browser, Pixotter's background remover handles most subjects without uploading a single byte.

Photoshop: The Gold Standard for Photo Cutouts

Photoshop's Select Subject tool (introduced in v21, significantly improved through v26.3) uses Adobe Sensei to detect and select the primary subject in one click. The real power comes from pairing it with Refine Edge Brush for hair and fine details.

Photoshop Cutout Workflow

1. Open your image. Select Object Selection Tool (W) from the toolbar.
2. Click Select Subject in the options bar. Photoshop draws the initial selection.
3. Enter Select and Mask workspace (Select > Select and Mask).
4. Switch to Refine Edge Brush and paint over hair, fur, and wispy edges. The algorithm distinguishes foreground strands from the background.
5. Set Output To: New Layer with Layer Mask and adjust Decontaminate Colors if background color bleeds into the edges.
6. Export as PNG with transparency: File > Export > Export As > PNG with Transparency checked.

Handling Hair and Fur

Hair is the hardest part of any photo cutout. Individual strands are semi-transparent, overlap unpredictably, and often share similar colors with the background. Photoshop's Refine Edge handles this better than anything else because it analyzes edge pixels probabilistically rather than drawing a hard boundary.

Tips for hair cutouts:

• Shoot against a contrasting background when possible. A blonde model against a beige wall is a nightmare for any tool.
• Use Refine Edge Brush at 50-70% radius — too small misses strands, too large smears the boundary.
• Enable Decontaminate Colors at 50-70% strength to remove color fringing from the old background.
• Check your mask against multiple backgrounds — toggle between black, white, and a colored background to spot missed areas.

GIMP: Free Photo Cutout with Foreground Select

GIMP's Foreground Select tool (SIOX algorithm) provides a semi-automated cutout workflow. It is not as refined as Photoshop's AI-powered selection, but it handles clean-edged subjects well and costs nothing.

GIMP Cutout Workflow (v2.10.38)

1. Open the image. Select Foreground Select Tool from the toolbox.
2. Draw a rough lasso around the subject — this defines the region of interest.
3. Press Enter. GIMP highlights the estimated background in blue.
4. Paint over the foreground subject with the brush. You do not need to be precise — broad strokes across distinct regions of the subject are enough.
5. Press Enter again. GIMP refines the selection.
6. Convert to a selection: Select > By Color > From Path (or use the generated selection directly).
7. Add an alpha channel (Layer > Transparency > Add Alpha Channel), invert the selection (Select > Invert), and delete the background.
8. Export as PNG: File > Export As > choose PNG format.

GIMP requires more manual work than Photoshop but produces clean results on subjects with well-defined edges — products, buildings, objects on solid backgrounds. For hair and fur, it struggles. Consider combining GIMP's selection with manual path tool refinements for challenging edges.

Canva: One-Click Photo Cutout for Non-Designers

Canva's Background Remover is a single-click tool that works inside the editor. Upload an image, click "BG Remover" in the effects panel, and Canva isolates the subject. The AI handles most common scenarios — people, products, animals — but does not offer edge refinement controls.

Best suited for social media graphics and quick collages where pixel-perfect edges are less critical. If you need to change the image background after cutting out the subject, Canva makes that part effortless too.

Limitation: Background Remover requires Canva Pro ($12.99/mo). The free tier does not include it.

Remove.bg API: Batch Photo Cutouts at Scale

When you need to process hundreds or thousands of images, a GUI tool is impractical. Remove.bg offers a REST API that accepts an image and returns the subject with a transparent background. The AI model is specifically trained on foreground/background separation and handles hair, pets, and products at quality comparable to Photoshop.

Remove.bg API Example

# Remove.bg API v1.0 — requires API key from remove.bg/api
curl -s -X POST "https://api.remove.bg/v1.0/removebg" \
  -H "X-Api-Key: YOUR_API_KEY" \
  -F "image_file=@input.jpg" \
  -F "size=auto" \
  -o cutout.png

Batch Processing with Remove.bg

#!/usr/bin/env bash
# Batch photo cutout using Remove.bg API v1.0
# Processes all JPGs in a directory

API_KEY="YOUR_API_KEY"
INPUT_DIR="./photos"
OUTPUT_DIR="./cutouts"
mkdir -p "$OUTPUT_DIR"

for img in "$INPUT_DIR"/*.jpg; do
  filename=$(basename "$img" .jpg)
  curl -s -X POST "https://api.remove.bg/v1.0/removebg" \
    -H "X-Api-Key: $API_KEY" \
    -F "image_file=@$img" \
    -F "size=auto" \
    -o "$OUTPUT_DIR/${filename}_cutout.png"
  echo "Processed: $filename"
  sleep 1  # Respect rate limits
done

Pricing: 1 free credit on signup. After that, $0.20/image on the pay-as-you-go plan or $0.07/image on the subscription plan (1,000 credits/mo for $69). For high-volume work, the subscription pays for itself quickly.

Python rembg: Self-Hosted Photo Cutout

If you want full control over the model, no per-image cost, and the ability to run cutouts entirely offline, rembg is the open-source answer. It uses the U2Net model for salient object detection and runs locally.

Installation and Usage (rembg v2.0.57)

# Requires Python 3.9+
pip install rembg==2.0.57 onnxruntime==1.17.1

# Single image cutout
rembg i input.jpg cutout.png

# Batch processing — entire directory
rembg p ./input_photos/ ./cutout_photos/

# Use a specific model (isnet-general-use for general subjects)
rembg i -m isnet-general-use input.jpg cutout.png

Python Script for Custom Pipelines

# rembg v2.0.57 with Pillow v10.3.0
from rembg import remove
from PIL import Image
import io

input_image = Image.open("product.jpg")
output_image = remove(input_image)
output_image.save("product_cutout.png", format="PNG")

rembg supports multiple models. The default u2net works well for general subjects. isnet-general-use often produces cleaner edges on detailed objects. u2net_cloth_seg is specialized for clothing segmentation — useful for fashion e-commerce.

After extracting your subject, you may want to make the PNG background fully transparent or compress the resulting PNG to reduce file size before publishing.

ImageMagick: Scripted Cutouts for Simple Backgrounds

ImageMagick is not an AI tool — it works on color math. For subjects photographed against a consistent solid background (product shots on white, headshots on green screen), it can extract subjects efficiently in a batch pipeline.

ImageMagick Photo Cutout (v7.1.1-38)

# Remove a white background (±15% color tolerance)
magick input.jpg -fuzz 15% -transparent white cutout.png

# Remove a green screen background
magick input.jpg -fuzz 20% -transparent "#00FF00" cutout.png

# Trim transparent padding after cutout
magick cutout.png -trim +repage trimmed_cutout.png

When ImageMagick works: solid-color backgrounds, consistent lighting, hard-edged subjects. When it fails: natural scenes, gradients, shadows, hair. For complex backgrounds, use rembg or Remove.bg instead.

Pixotter: Browser-Based Photo Cutout with Zero Upload

Pixotter's background remover runs entirely in your browser using WebAssembly. Your images never leave your device — the AI model processes everything client-side. This matters for sensitive photos (employee headshots, medical images, legal documents with photos) where uploading to a third-party server is not an option.

The workflow: drop your image, click remove background, download the cutout as a transparent PNG. You can then resize the result for your target platform or convert it to another format — all in the same session without re-uploading.

Choosing the Right Output Format

After cutting out your subject, the format you export to matters:

PNG is the safe default — every platform accepts it. If the cutout is destined for a website and you want to save bandwidth, convert to WebP or AVIF. Never save a photo cutout as JPEG; the white or black background fill defeats the entire purpose.

Common Photo Cutout Use Cases

Product Photography

E-commerce platforms like Amazon, Shopify, and Etsy require or strongly prefer product images on pure white backgrounds. A clean cutout placed on a white canvas meets listing requirements and looks professional. Use Remove.bg or rembg for batch processing entire product catalogs.

Profile Pictures and Headshots

Corporate headshots, Slack avatars, and LinkedIn photos often need the background replaced or removed. Cut out the subject, then either keep the transparent background or place it on a branded color. For passport photos with specific background color requirements, a cutout is the first step.

Compositing and Collages

Designers frequently extract subjects from multiple photos to combine them into a single scene. Precise edge quality matters here — any fringing or halo from the original background is visible when the subject sits on a new background. Photoshop or rembg with post-processing produces the cleanest composites.

Social Media Graphics

Instagram carousels, YouTube thumbnails, and LinkedIn posts often use cutout subjects layered over branded backgrounds or gradients. Canva and Pixotter are the fastest paths here since you can do the cutout and the design work in one place.

Tips for Better Photo Cutouts

1. Start with good source photos. Sharp focus on the subject, contrasting background, even lighting. A great tool cannot save a blurry subject shot against a similarly-colored background.
2. Check edges at 200% zoom. Haloing (a faint outline from the old background) is the most common cutout defect and is invisible at normal zoom.
3. Use Decontaminate Colors in Photoshop or post-process edges in any tool. Color spill from the original background bleeds into edge pixels.
4. Export at the original resolution. Cutting out from a downscaled image produces noticeably worse edges. Work at full resolution and resize afterward.
5. Test against multiple backgrounds. A cutout that looks perfect on white might reveal edge artifacts on dark backgrounds. Toggle between black, white, and a mid-tone color.
6. For batch jobs, sample first. Run 5-10 images through your pipeline before processing the full set. Spot-check results, adjust parameters (fuzz values, model selection), then process at scale.

Frequently Asked Questions

What is the best free tool for photo cutouts?

For most people, Pixotter's background remover is the best free option — it runs in your browser, handles complex subjects well, and never uploads your images. For developers who want CLI access, rembg (v2.0.57) is free, open-source (MIT license), and runs entirely offline.

How do I cut out a subject with hair or fur?

Photoshop's Refine Edge Brush is the best tool for hair. Paint over the hair boundary in Select and Mask mode, enable Decontaminate Colors, and export with a layer mask. For free alternatives, rembg's isnet-general-use model produces decent hair edges, though manual cleanup in an editor may still be needed.

Can I batch process photo cutouts?

Yes. Remove.bg's API, rembg's rembg p command, and ImageMagick all support batch processing. Remove.bg charges per image; rembg and ImageMagick are free. For simple solid-color backgrounds, ImageMagick is fastest. For complex natural backgrounds, rembg or Remove.bg produce usable results without manual intervention.

What format should I save my photo cutout in?

PNG is the universal safe choice — it preserves full transparency and every platform accepts it. For web publishing, WebP offers 25-35% smaller file sizes with the same alpha channel support. Avoid JPEG entirely for cutouts since it does not support transparency.

How do I remove the white edges (halo) around my cutout?

White halos appear when edge pixels retain color from the original background. In Photoshop, use Decontaminate Colors in Select and Mask at 50-70% strength. In GIMP, use Filters > Light and Shadow > Erode on the alpha channel by 1 pixel. Programmatically, apply a 1px alpha erosion using Pillow or ImageMagick's -morphology Erode Disk:1 operator.

Is it safe to use online photo cutout tools?

It depends on the tool. Cloud-based services like Remove.bg and Canva upload your image to their servers for processing. If you are working with sensitive images, use a client-side tool like Pixotter (processes entirely in your browser) or a local tool like rembg (runs on your machine). Your images never leave your device with either option.

How do I cut out a subject and put it on a new background?

First, extract the subject using any tool in this guide. Save as a transparent PNG. Then place the cutout on your new background in your editor of choice — Photoshop, GIMP, Canva, or even CSS if it is for a webpage. For a complete walkthrough, see our guide on how to change an image background.

Can I extract multiple subjects from one photo?

Most AI-based tools (Remove.bg, Pixotter, rembg) detect and extract all foreground subjects together. To extract individual subjects separately, use Photoshop's Object Selection Tool — click each subject individually to create separate selections, then export each to its own layer and PNG. For removing specific people while keeping others, Photoshop's Generative Fill is the most reliable approach.