How to remove a hand from a photo
A hand reaching in from the left edge of a group shot. Someone's fingers still visible after you cropped out the rest of them. A hand holding a prop that stayed in frame when everything else got cut. Hands are small, but they draw the eye immediately when they shouldn't be there.
Why hands are tricky to remove from photos
Hands are harder than most objects to remove for a few reasons. First, skin tone blends. A hand that overlaps with a person's arm, neck, or face shares similar color values, which makes it harder for detection to separate one from the other. The algorithm sees flesh-colored shapes and has to infer where one ends and another begins.
Second, hands tend to touch things. A hand holding an object means the removal has to reconstruct both the background behind the hand and the portion of the object that was obscured. If someone is holding a glass at a party and you want to remove the hand, BOARD has to fill in both the table area behind the hand and the part of the glass the hand was covering.
Third, hands are structurally complex. Five fingers in various positions do not form a clean silhouette. A closed fist is easier than an open hand with spread fingers, where the gaps between fingers mean the fill has to work around irregular edges.
None of this makes hand removal impossible. It means the result quality varies more than it does for simpler objects like a trash can or a car.
When removing a hand from a photo works well
Good candidates
- A hand or arm entering from the edge of the frame, disconnected from the main subject
- Fingers visible at the border of a cropped photo
- A hand holding an object against a simple or plain background
- A hand resting on a table or surface where the area underneath is a single color
- Someone's arm in the far background of a wide shot
Harder cases
- A hand resting on someone's shoulder where the shoulder must be reconstructed
- Fingers intertwined with the subject's clothing or hair
- A hand in front of someone's face
- Multiple hands from different people that overlap each other
- A hand holding something colorful or complex against a matching background
Remove a hand from a photo: four steps
Go to app.brd.ing and upload the image. Original files produce better results than compressed copies. If the hand is at the edge of the frame, the fill only needs to reconstruct the background area, which is typically the easier case.
BOARD identifies all detectable objects in the scene. Hands, arms, and people are outlined as selectable elements. Hover over the photo to see what has been detected. If the hand is part of a larger person detection, you may see the whole person outlined rather than just the hand. That is fine for the next step.
Click the hand. If BOARD has detected it separately, you will see the hand outline. If it is grouped with a larger person detection, click the closest detected region. Run removal. The software fills the area where the hand was using the surrounding context.
Use the compare slider to look at the before and after. Pay attention to where the fingers were, especially if they were spread. The fill between fingers is the most common place for artifacts. If there is a smear or color mismatch, run a second pass on that specific area. Download when the result looks clean.
After removing a hand from a shoulder, check that area for residual skin-tone bleed. The boundary between the removed hand and the kept shoulder often needs a second pass at higher zoom to look clean at full resolution.
Frequently asked
Can AI remove a hand from a photo?
Yes, for hands that are clearly separate from the main subject. A hand at the edge of the frame, a hand holding something against a simple background, or a stray arm from outside the frame are all good candidates for clean removal. Hands intertwined with the subject are harder and may leave artifacts, but a second pass usually improves the result.
What if the hand is on someone's shoulder?
This is one of the harder cases because removing the hand means reconstructing the shoulder underneath it. BOARD can handle this, but the fill at the skin boundary may not be perfectly clean on the first pass. Zoom in after the removal and run a second pass on any area where the shoulder edge looks unnatural. Most of the time this produces a good result.
Does removing a hand use one credit?
Yes. Each removal operation uses one credit regardless of what you are removing. The first five edits are free with no account required. After that, each removal is $0.50.
Remove the hand from the photo.
Five edits free. No account, no app to install. See the result in seconds.
Clean Up a Photo Free →