> For the complete documentation index, see [llms.txt](https://cartoonmango-1.gitbook.io/vida-ekyc-integration/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://cartoonmango-1.gitbook.io/vida-ekyc-integration/product-overview/liveness-detection.md).

# Liveness Detection

Liveness detection is an important component in the digital signature issuance process, as it helps verifying the authenticity of an individual's identity. VIDA leverages advanced AI and ML algorithms to create a passive liveness method that analyses a physically present human being and distinguishes it from a spoof artifact that may be present in a single-shot image. This method improves your conversion by having less friction in the capture experience and ensures that the person undergoing verification is physically present during the process and not using a pre-recorded image or video.

The purpose of liveness detection is to prevent fraud and impersonation, providing a higher level of security and protection to the verification process. It also ensures that only legitimate and authentic users are able to obtain digital signatures.

<figure><img src="https://272753921-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FwexkayGypqDBWfSUkBJE%2Fuploads%2FSL0RqGbWHSJJ5veChjMD%2FLiveness%20flow.jpg?alt=media&amp;token=4874fe6e-8c91-4a78-906e-84d3a405cf1e" alt=""><figcaption></figcaption></figure>

**Passive Liveness Check**

Passive liveness check verifies the authenticity of a selfie in the backend, by analysing various aspects of the selfie to determine if the person in the picture is actually present and not a fake or manipulated image. This is accomplished through the use of advanced image analysis techniques using AI and ML.

VIDA offers API to check the liveness of the captured selfie in the backend. VIDA system checks if the provided selfie is live/real or hacked along with a few additional image checks and generates a selfie score.

The selfie score ranges from 0 to 1. The score of 0 (zero) represents maximum liveness and the score of 1 represents minimum liveness of the selfie. However, the success and failure of the liveness check depend upon the liveness threshold score.

{% hint style="info" %}
**Liveness detection** can be **disabled** on requirement. Please contact the [<mark style="color:blue;">VIDA support team</mark>](mailto:support@vida.id)<mark style="color:blue;">,</mark> incase if Liveness detection needs to be disabled.
{% endhint %}

##

## Image Type and Resolution&#x20;

The format and resolution of the image captured for liveness detection are important factors that affect its ability to be processed and analysed accurately by the system.

Ensuring that the image is in the correct format and has sufficient resolution is necessary to obtain an accurate liveness detection result.

| Characteristic                             | Specification                                                                                                                     |
| ------------------------------------------ | --------------------------------------------------------------------------------------------------------------------------------- |
| Image Resolution                           | <ul><li>Minimum : 300*400 </li><li>Maximum : 3000*4000</li></ul>                                                                  |
| Image Size                                 | <ul><li>Minimum : 100KB </li><li>Maximum : 1MB</li></ul>                                                                          |
| Image Type                                 | JPEG                                                                                                                              |
| Image file size Compression recommendation | Bicubic compression or resize the image resolution without affecting the image quality                                            |
| JPEG quality                               | No less than 80%                                                                                                                  |
| Image aspect ration                        | <p>use the similarity of the initial ratio to reduce the size.</p><ul><li>Minimum ratio 9:16</li><li>Maximum ratio 16:9</li></ul> |
| Selfie Orientation                         | Frontal face, portrait straight to the camera and not rotated                                                                     |
| Face area size                             | <ul><li>Minimum face area 160 x 160 px</li><li>Maximum face area respective to the maximum allowed image file size</li></ul>      |

## Recommendations for Successful Liveness Detection

### Quality Check of Capture Selfie

The quality of the image captured for liveness detection can impact the accuracy of the result.

A well-lit, clearly visible image with a neutral background is necessary for the system to accurately analyse the face.

* User should ensure that the image is well-lit and not too dark or too bright
* User should avoid taking the selfie in front of a bright light source
* User should ensure that the background is plain and neutral, without any distracting elements
* User should avoid taking the selfie which produce glare
* User should avoid taking the blur selfie

### Guidelines for Optimal Images

| Image                                                                                                                                                                                                                                                    | Image Condition    | Acceptance/Rejection                                                   |
| -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------ | ---------------------------------------------------------------------- |
| ![](https://lh4.googleusercontent.com/baXq7Ypuh6CnspfdMD61NQV_O32M-DvFx9sCzMN03GbWI7YkThfyhqTiB2YgfLWuMtzA2GpJq0O_ZkdK5eaUGllTj-nVcjYkDRw92XFHl5FRExXjFFy4seAm-2VYT5B4-zFW-37OoBKoaETmlHg8f-E)                                                           | Ideal Condition    | **Highly recommended** for optimal OCR accuracy                        |
| ![](https://lh3.googleusercontent.com/2RwoT1uQw-wnHj0YNwOwTpjbCYKoaDxBGXydX60alZBPoCZxNhl1pB-pFrhkDQXOmg6c8HdsyG_nOIkjHFnX3fZaDXQ_JLI_hbxCV9qiWyWDIn3SFn5IQIzzdnTo8KLZsPORMuOgEPN26O7wCt64WJ8)                                                           | Relatively Safe    | **Acceptable**, but may impact OCR accuracy                            |
| ![](https://272753921-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FwexkayGypqDBWfSUkBJE%2Fuploads%2FS63RknWkHhGPbTtCASaU%2FScreenshot%202023-04-12%20at%209.15.50%20PM.png?alt=media\&token=3b8eb874-aafe-4272-9a50-503f4bd3a308) | Warn to be Avoided | **Potential to be rejected** due to significant impact on OCR accuracy |

### Guidelines for Defective Images

<table><thead><tr><th>Image</th><th width="154">Image Condition</th><th width="176">Acceptance/Rejection</th><th>Recommendation</th></tr></thead><tbody><tr><td><img src="https://lh4.googleusercontent.com/2wIxMy4-5ZdGwCnxvSIFv_RDFF0BSyc-4MXE87IyO69HeN-GWBylY5bX6fp0pIapEglAxcXvXi4x8jrEN1UPoQ2CAzdg0IR3-paaS_6pI6v8G8zBSh_vQ8dYEx1CF6fE-MoOLVFOR2HHGbPOom7ohpo" alt=""></td><td>Low Light</td><td><strong>Should be avoided</strong> as it can significantly impact OCR accuracy</td><td>The image should be well-lit and properly exposed</td></tr><tr><td><img src="https://272753921-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FwexkayGypqDBWfSUkBJE%2Fuploads%2F8DSqTPWfvYYb7GyKh1Nk%2FScreenshot%202023-04-12%20at%209.31.39%20PM.png?alt=media&amp;token=7c464b09-21e7-40cb-8422-5a2fc7f8bed9" alt=""></td><td>Blur</td><td><strong>Should be avoided</strong> as it can significantly impact OCR accuracy</td><td>The image should be sharp and in focus</td></tr><tr><td><img src="https://lh3.googleusercontent.com/cXGiyPDcFSZnVsLkXgP-JaRilFqA7A6AbvyOK0BqgI3qDFzCoecjLD-p0pTgcQfR0FHF1nkkRgj-dWFp7-Hfq2Nq5jTEPOkFgDJ8vo59cQB2xiDNX5IPUk9rK8I-RO2bWmEcZgp_YW7UBv1nRVDJM9k" alt=""></td><td>Defect/Corrupt</td><td><strong>Should be avoided</strong> as it can significantly impact OCR accuracy</td><td>The image should be clean and undamaged</td></tr><tr><td><img src="https://lh3.googleusercontent.com/t9hL9zAj-OcS6nW-n7hx2_Cc5w6MZEZ9zlPuGUwjRPZGGPlEaDVszP_1EuoFlpr6EdToS1kBaRY3FLhJ4CiqZCXLfkZk6ztH6mHnWoE0gy771u4SpYaxQknqSG7nEWbxRIVmlfZ_9uAgmbnv5KGwuBg" alt=""></td><td>Over-exposure</td><td><strong>Should be avoided</strong> as it can significantly impact OCR accuracy</td><td>The image should be properly exposed with no blown-out highlights</td></tr><tr><td><img src="https://lh4.googleusercontent.com/lfTzC8wR2yj7-_qQFDBnVmbpFGYaokoZasDGy1-NwidRUJQ8Qm2jD0nqA17pVtd5WE5lutKvrJVE-J6G0oA7V1VXrV4w6g7XKuuVJAMCydrsu9hQuyU5kHA89o-R9Imzv5rgLc_5Fieo1K2nePm1afo" alt=""></td><td>Low Contrast</td><td><strong>Should be avoided</strong> as it can significantly impact OCR accuracy</td><td>The image should have good contrast with clearly defined edges</td></tr><tr><td><img src="https://lh3.googleusercontent.com/sITxu_DcW0NfW5QIMQzB99i1BOMWS3ymopt2QOha1fdyInc5Baw7O5c5VaQNcZvNaTnpw7Fo1WNgz3bQLd_-oAae3EDGPMNHfHC23QjJp9MOkVswtAC0e911GjAy0RSkaTC6Un4rUN1RfqtsXopxq20" alt=""></td><td>Glare</td><td><strong>Should be avoided</strong> as it can significantly impact OCR accuracy</td><td>The image should be free of glare and reflections</td></tr></tbody></table>

### Face Appearance Guidelines

The appearance of the face in the image captured for liveness detection can impact the accuracy of the result.

* User should make sure that their entire face is visible and not partially obscured by hair, glasses, mask or any other objects (occlusion)
* Keep a neutral expression, avoiding smiling, frowning, or making any other exaggerated facial expressions
* User should ensure that their face is straight-on and positioned in the centre of the frame.
* User should avoid selfie-holding KTP. When holding a KTP/ID card, the condition may be detected as spoof activity.

<table><thead><tr><th>Image</th><th width="161">Condition</th><th>Reason</th><th>Recommendation</th></tr></thead><tbody><tr><td><img src="https://lh6.googleusercontent.com/WNOvNraxKwXO_sWMHunX2vmDTPHdXyFJEYOBGh3TJKIpyDmz4cyNyGueGKtMthXYIs4O5VLh-qiLDSiZoYfP1haXr0Fxq68BWCnwc__ERm73ZViRYEmgsCdIqXXRf98TOztBdI4ty1H0e3rK0muw8EU" alt=""></td><td>Ideal Condition</td><td>NA</td><td>NA</td></tr><tr><td><img src="https://lh6.googleusercontent.com/AovzGT69qML1xoC_p6booGby97PJCQFu0aKjQ_gMjmIz8fNeQAh22ZTGGZpF04lboXXZFF_tAYcFbX9_pJ0KTytmE4KQM3eUtLci9w6GlPjHgxSgcINyvVdID5tXXYCvOCMR4t1TOn80N4xmJP5IQAs" alt=""></td><td>Ideal Condition</td><td>NA</td><td>NA</td></tr><tr><td><img src="https://lh5.googleusercontent.com/HzpBLfe72YDD3rZsiSCIqo09XeaVj_J7F2SA64Ke_PePPRfSOaT2IZZ2mo-mJ-EexnAr7Pn2aoQunqWgNQpqUl_klLUw7Y6_R6DbdtZCdK7FtFOxe6qNx_9qNHSt3qMxhcDKVB1mXoLXUnPkPIPwf9s" alt=""></td><td>Ideal Condition</td><td>NA</td><td>NA</td></tr><tr><td><img src="https://lh5.googleusercontent.com/blm7VryPdZeSDNpvhxmRkf27zHwkm2hdsOVxOCt49j2Ib1cpVYnE8Dtw0CDPSGf3GplW6117mT0eCR5Joe2ya7T0cwamBz2iM650JaIs-OhrAYEU2TZWsNjn9k7Cx_415ZWnYV4_WUKJ61DCCoHNFxw" alt=""></td><td>Ideal Condition</td><td>NA</td><td>NA</td></tr><tr><td><img src="https://lh5.googleusercontent.com/TA9S1sPks6unMUwwKKIIkJmGq9mKyXApprHCpmAus8Y83__29EkC1NNoZwMR5tHyFdm36L5xVqn9CvGDQPvPLX9j6PsY-GdVtpncFEXQr1sJUMyHOqXTW_v4_gU3B54jMkjFRJSkBYqTbN_HLDSKPeQ" alt=""></td><td>Ideal Condition</td><td>NA</td><td>NA</td></tr><tr><td><img src="https://lh4.googleusercontent.com/zuh2QsH7LDWG3_8H8FA0yuf1kxif-SWKqoS74Gjb9gYtLIvp4jqNVZiI2SFb8xUxzv2j_YbOkte1a5DIrDWUCJROEkrhfXDII8dOw1lB3cTHby-ODF1yzm2XVV8XW4QehvCsLsH3WxXpTmtftWLTNfs" alt=""></td><td>Ideal Condition</td><td>NA</td><td>NA</td></tr><tr><td><img src="https://lh5.googleusercontent.com/gpuSgPQU0xwoTi9zVbn2O5F1LQ7mYW976djxsfmlNcBxji52x1k6HUVhyPS26pXwePrQCa_FSPMdxXhvkcZC77jtv4nMhto4cMIiaOYABjKVIzaa1QBw6N5d4RrJegrHLgpxtxp5fxEPeYNaRhgySOk" alt=""></td><td>Warn to be Avoided</td><td>Wearing specks that are not properly positioned and covering the eyes</td><td>Warning, suggest retaking photo</td></tr><tr><td><img src="https://lh3.googleusercontent.com/v0VSCPLBio-hLWOtPn-97Usbl0dzXC0dyMWG8rpgw_RX9SVadzDzJR6iuv2MCPUdccf3FofCkbbv1oX4wJ6VcIFZ2eHN7_5mPEufnQBEqg-GD9YV0s4rUd2AA65En1enKhFhRBHy7fvdYC8Nc-XF_rQ" alt=""></td><td>Warn to be Avoided</td><td>Hair obstructing part of the face and eyes</td><td>Warning, suggest retaking photo with hair pulled back</td></tr><tr><td><img src="https://lh5.googleusercontent.com/j3sysoLnmxu7sMJFf00-J-rE6PsT0jkUdZKHL7pQ_C_p3-HBkJi7_jV-McByvMdb0rB73AayLsqzymiZ79Ny3hTjgLawUTjbyIxvBWXCSShgjz5n3BE8m5j48RZAIMtizriql-KfFwZzomRXHSbXWP8" alt=""></td><td>Warn to be Avoided</td><td>Face partially cut off in the image</td><td>Warning, suggest retaking photo with entire face visible</td></tr><tr><td><img src="https://lh4.googleusercontent.com/HLOh7RBWICoXWdE7ELhkiO-gZC-j-HeRJcYy7f-jzQ_Q0gvg2hcUWLonGciTXGTHpfpN-9zy4826Y8K1ss7KhORaLpAgAttFr9HhWtlXV0qauuMRHZ2qxm3h162kU_3goQHZqIlB5Xbrgqx8ZvSBOMQ" alt=""></td><td>Rejected by the system</td><td>Hair completely covering the face and eyes</td><td>Rejection, suggest retaking photo with hair pulled back</td></tr><tr><td><img src="https://lh6.googleusercontent.com/LlMa3bEdp_j1Iqk7Ou8M5oPYPjyCWSyo6zsFWwd7m0Mxjrf06hr7wwENMbggx9L2nBwAvhMWTLsEbDE6-Qpr5nMuygXEZ_slbjl8MopkgR6iw7mhPRjvMR-54b_unDEEYFoKnxhv6-Qijn6SRROxbHY" alt=""></td><td>Rejected by the system</td><td>Wearing colored cooling glasses that obscure the eyes</td><td>Rejection, suggest retaking photo with clear glasses</td></tr><tr><td><img src="https://lh5.googleusercontent.com/FtOiByPT9ZoE0VlqzOzHRkCQgnxU0NmfxUT8Mr3zR7FQir5FragK6qSOO-RxWAbPC0s1gfxOLZ3sNg1V9KKsLBDjOcgq1WQp-nUPQNSAMe5Mtz0zyvg-cfBbI5LPsEmjRxF9EhTLvxjetwSXxXnDSHs" alt=""></td><td>Rejected by the system</td><td>Wearing a mask that covers a significant portion of the face</td><td>Rejection, suggest retaking photo without a mask</td></tr><tr><td><img src="https://lh3.googleusercontent.com/ESJ4oOtscVBn1qArj9A-0j5HDFMpztc-5kaOYJC3RXR3SDEV_ea75ZSlgtGYaxVAOvZDuoKnn7Vv06iz5jxERt_hjuWyOakQnvVspb3gaU-wZpdSopDKu17dS3BaCVPhcSOOlrzhf5ey43OJvkmXwGE" alt=""></td><td>Rejected by the system</td><td>Wearing a hat that obstructs the view of the eyes</td><td>Rejection, suggest retaking photo without a hat</td></tr><tr><td><img src="https://lh4.googleusercontent.com/ma2NAqwgi5X41o-ZM2pwq38KaJ4Owqq8BGJYOJ5K5Bzdi2hQurw4iy_aCrfrfiLXrCBYROPYcCOywUyOomhxHvisAYCXcNr8kFsGTcICu_ElwOeGg_mLmuIVA5Gg8Dr1ovfebVId8Bf7AORYY5CAe40" alt=""></td><td>Rejected by the system</td><td>Hand covering part of the face</td><td>Rejection, suggest retaking photo with hands away from face</td></tr><tr><td><img src="https://lh6.googleusercontent.com/UTCkKjjK_cUo4-gtMpLX3WJDQhArHOJZ5U_okc8GYqB4ZhQi7EicIPYlz3acqAy7MX1NozvnVqRLDcpAQRLjlU3nWjxOuXb61i8GexNTXTD-iN6Nqk2c4QL017_ND9v7Fja8a6Wdqh_nYRg_np1VX2Y" alt=""></td><td>Rejected by the system</td><td>Face not oriented properly or at an angle</td><td>Rejection, suggest retaking photo with face oriented straight ahead</td></tr><tr><td><img src="https://lh4.googleusercontent.com/V1QGcytlwihBTyH54t7OvLhC5WPtjGkDhuhFrkgO7S5L3PCyQLLt77dqPvJiki9DK3bEoygEjygjl_pEuAT64n-SY1CZevkpFcVPSqh378Os_jDtwhQY3oWf7RArhqUd9WLE4tiELHnuXkhjNMVMtmk" alt=""></td><td>Rejected by the system</td><td>User holding KTP in the selfie</td><td>Avoid holding KTP </td></tr></tbody></table>

<figure><img src="https://272753921-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FwexkayGypqDBWfSUkBJE%2Fuploads%2Fp58mLmaM63L28PVHsywq%2FLiveness%20Error.jpg?alt=media&amp;token=ff403ede-302c-4b73-9eec-9078761640fe" alt=""><figcaption></figcaption></figure>