Free Tool
AI depth-of-field blur from any portrait photo
Depth-estimated background blur that keeps your subject sharp — no DSLR, no Photoshop.
The problem
Achieving a shallow depth-of-field background blur on a phone camera requires portrait mode, which is limited by sensor size and often blurs subject edges unnaturally. Replicating the effect in Photoshop requires manual selection and masking, which takes significant skill and time.
The NSS solution
NSS Portrait Blur uses a depth-estimation model to identify foreground and background regions. Pixels with depth below 0.15 (near the camera) are kept sharp. Everything else receives a Gaussian blur scaled by the strength slider. The result is a natural, graduated blur that follows the scene's actual depth.
How to use it
- 1
Add background blur to phone portraits
Phone cameras without portrait mode produce flat, in-focus backgrounds. Upload the photo and the depth model adds a natural-looking background blur in seconds.
- 2
Strengthen or weaken existing portrait mode blur
Upload a photo shot in portrait mode and apply additional blur to make the background recede further, or reduce the effect if the original was too aggressive.
- 3
Professional headshots from any camera
Portrait mode is not available on many webcams or entry-level cameras. NSS Portrait Blur gives those headshots a professional shallow-DoF look without re-shooting.
- 4
Product photography with blurred context
Place a product in a real environment, shoot it, then apply background blur to keep focus on the product while adding environmental context.
- 5
Social media profile photos
Turn a casual snapshot into a professional-looking portrait for LinkedIn, Twitter, or Instagram profile photos with a clean, blurred background.
Step-by-step guide
- 1
Upload your portrait
Go to /portrait-blur and drop your image. Works best with a clear foreground subject and a background at a different distance from the camera.
- 2
Wait for depth estimation
The depth model analyses the scene and builds a depth map. This takes 2–8 seconds depending on image size and whether WebGPU is available.
- 3
Adjust the blur strength slider
Start at 50% and move the slider to increase or decrease blur intensity. The preview updates in real time.
- 4
Check foreground sharpness
The foreground protection threshold (depth < 0.15) keeps your subject sharp automatically. If the subject looks blurry, the depth model may have misclassified a region — try a photo with more subject-background contrast.
- 5
Download your image
Click Download to save the result as a PNG at the original resolution.
Common questions
Does it work on group shots?
Yes, but results vary. The depth model treats the closest subjects as foreground. If multiple people are at very different distances, some may be partially blurred. Works best on single or two-person portraits.
Can I use photos taken on a phone camera?
Absolutely. Phone photos are the primary use case. Images shot at 12 MP or higher give the best results. HEIC files are automatically converted in the browser.
How does phone camera blur compare to DSLR results?
DSLR lenses produce optical blur that varies by aperture and focal length. NSS Portrait Blur simulates this with a Gaussian model driven by depth estimation. The result is visually similar for most portrait shots, though fine bokeh balls and lens-specific renderings differ.
Is the depth estimation model downloaded each time?
The model is cached in your browser after the first download. Subsequent uses load instantly from cache without any network request.
Ready to try it?
No account. No subscription. No images uploaded to any server.
Add portrait blur free→