Why Does Binance Face Recognition Keep Failing?

The face recognition step in Binance's KYC verification causes headaches for many people — staring at the camera for a long time only to keep failing. Some have tried over a dozen times and are ready to give up. In fact, face recognition failures usually have specific reasons, and addressing them accordingly will solve the problem. This article will explain the various reasons for failure and their solutions. You can do face recognition on the official Binance website, but it is highly recommended to do it on the official Binance App, as using the front camera of a mobile phone provides a better experience. Apple users should refer to the iOS installation guide to install the App.

How Face Recognition Works

Before analyzing the reasons for failure, let's first understand how Binance face recognition works:

1. Liveness Detection: The system will ask you to perform certain actions (blink, open your mouth, turn your head, etc.) to confirm that there is a real person in front of the camera, not a photo or video.
2. Feature Extraction: The system extracts key facial feature points from your real-time facial image.
3. Comparison and Matching: It compares the extracted facial features with the ID photo you uploaded.
4. Comprehensive Judgment: Based on the liveness detection results and the matching degree, it gives a pass or fail judgment.

A problem at any of these steps can lead to ultimate failure.

Common Reasons for Face Recognition Failure

Lighting Issues

Lighting is the most significant factor affecting face recognition.

Too Dark: Facial images captured by the camera in low light conditions have a lot of noise and few details, making it impossible for the system to extract facial features accurately.

Too Bright/Backlit: Facing a strong light source (like a window or a bright bulb) will cause the face to be overexposed and turn completely white, losing all features. Alternatively, backlighting can turn the face into a silhouette.

Uneven Lighting: Having half of the face bright and the other half dark creates asymmetrical facial features, affecting system recognition.

Strobing Lights: Some fluorescent lights produce a strobe effect, which can cause alternating light and dark in the video, interfering with recognition.

Facial Obstructions

Masks: Wearing a mask covers the lower half of the face, preventing the system from capturing complete facial features.

Hats and Headscarves: These cover the forehead and hairline, affecting the recognition of facial contours.

Sunglasses or Dark Glasses: These obscure the eye area, and eyes are one of the most critical feature areas for facial recognition.

Bangs Covering Eyes: Excessively long bangs that cover the eyes or eyebrows.

Significant Differences from the ID Photo

Age Changes: If your ID photo was taken many years ago, your face might have changed significantly.

Weight Changes: Significant weight gain or loss can alter facial contours.

Makeup Differences: The difference between heavy makeup and a bare face can sometimes affect system matching.

Wearing Glasses: If you weren't wearing glasses in your ID photo but are wearing them now, or vice versa, it may decrease the matching degree.

Hairstyle Changes: Although the system primarily recognizes facial features, drastic hairstyle changes can occasionally have an impact.

Network and Device Issues

Unstable Network: Face recognition requires real-time video data transmission. Network lag can cause data transmission to be interrupted or delayed.

Poor Camera Quality: The cameras on very old phones or computers have resolutions that are too low to capture facial details clearly.

Obstructed or Dirty Camera: Fingerprints, dust, etc., on the camera lens affect image clarity.

Operational Issues

Uncooperative Actions: You didn't blink when the system asked you to, or you performed the wrong action.

Moving Too Fast or Too Slow: The system has a specific time window to detect actions. Too fast and it might miss it; too slow and it might time out.

Incorrect Distance: Your face is too close to or too far from the camera, so it cannot be completely captured or is too small to see clearly.

Constant Shaking: Constantly moving your head during face recognition prevents the system from stably capturing facial features.

Solutions for Face Recognition Failures

Optimize the Lighting Environment

This is the most important and effective improvement:

1. Face a Window: Sit facing a window to let natural light evenly illuminate your face.
2. Choose the Right Time: Natural daylight works best; avoid using a desk lamp at night if possible.
3. Avoid Overhead Lighting: Do not operate directly under overhead lights, as this easily creates shadows on the face.
4. Supplemental Lighting: If natural light is insufficient, place a desk lamp in front of you for fill light, but not too bright.
5. Even Illumination: Ensure the light on both sides of your face is generally even.

Remove Facial Obstructions

1. Take off your mask.
2. Remove hats and headscarves (except for religious reasons, in which case you can contact customer service for special handling).
3. If your glasses have severe glare, take them off.
4. Pin your bangs to the side to expose your entire face.
5. Remove heavy makeup and try to maintain a state similar to your ID photo.

Minimize Differences with the ID Photo

1. Refer to the ID Photo: Look at what you look like in your ID photo and try to match it.
2. Match Glasses Usage: Wear or don't wear glasses to match the ID photo.
3. Light Makeup or Bare Face: Keep it consistent with your ID photo.
4. Natural Expression: Maintain a calm, natural expression; do not force a smile or make faces.

Improve Network and Device

1. Connect to Stable WiFi: Switch to the WiFi network with the strongest signal.
2. Close Other Apps: Reduce the bandwidth usage of background apps on your phone.
3. Clean the Camera: Wipe the front camera lens of your phone with a soft cloth.
4. Use a Better Device: If your phone is too old, switch to a device with a better camera.

Correct Operating Method

1. Maintain Proper Distance: Keep a distance of 30-50 cm between your face and the camera.
2. Center Your Face: Position your face in the middle of the on-screen frame.
3. Keep Steady: Try not to move your head drastically when performing actions.
4. Follow Prompts: Blink when the system says blink, open your mouth when it says open your mouth. Do not jump the gun.
5. Appropriate Speed: Perform actions neither too fast nor too slow; just be natural.

Switch to the App

If you keep failing on the web version, it is strongly recommended to switch to the Binance App:

• The App interfaces more directly with the front camera.
• The App's face recognition algorithm might have additional optimizations.
• The App's network requests are usually more stable.

What to Do After Multiple Failures

Wait a Bit Before Retrying

If you have failed several times in a row, don't rush to keep trying. Take a break and try again at a different time (such as during the day when lighting is good). Continuous failures might cause the system to temporarily restrict you from retrying.

Change Document Type

If you keep failing face recognition using your ID card, try using your passport. Different ID photos have different shooting styles, and one might match your current appearance better.

Contact Customer Service for Manual Review

If you fail after many attempts:

1. Contact Binance online customer service or submit a ticket.
2. Explain that you have tried face recognition multiple times but failed.
3. Customer service may arrange for a manual review.
4. For manual review, you may need to submit:
   • A selfie holding your ID document.
   • A selfie holding a piece of paper with specific text (customer service will tell you what to write).
   • Other supporting documents.
5. Manual review requires some processing time.

Frequently Asked Questions

Can I do face recognition while wearing glasses?

Yes. However, if the lenses have severe glare that obscures your eyes, it is recommended to take them off. If you are not wearing glasses in your ID photo, it is best to take them off for face recognition to maintain consistency.

Will face recognition save my facial data?

Binance processes your biometric data in accordance with its privacy policy. The data is mainly used for identity verification purposes and is stored encrypted. Specific data retention policies can be found in Binance's privacy policy document.

Can twins pass face recognition?

Modern facial recognition technology can distinguish between most twins. As long as you are using your own ID document, just perform the face recognition normally. The system compares multiple facial feature points for precise identification.

Can I still pass face recognition if I have had plastic surgery?

If plastic surgery has caused significant changes to your facial features and your ID photo is from before the surgery, you may not pass automated face recognition. In this case, it is recommended to first update your ID photo (renew your ID card or passport) or contact Binance customer service to explain the situation and apply for manual review.

Is there a specific time of day when face recognition is more successful?

There is no specific time. However, the success rate of face recognition under daytime natural light is usually higher than at night. Just choose a time when it is convenient for you to find good lighting.