Web SDK
Last updated
Was this helpful?
Last updated
Was this helpful?
Published on 2025-05-14
Document Scanning
Added support for Moroccan ID (MAR_ID) scanning.
Added support for the latest Bahrain ID version.
New Document Types
Introduced QAT_ID_NATIONAL, QAT_ID_RESIDENT, KWT_ID_NATIONAL and KWT_ID_RESIDENT.
Note: Previous SDK versions reported types such as OMN_ID_NATIONAL simply as OMN_ID in the result object. From this release onward, the exact selected document type will be returned.
Tracing Enhancements (Analytics)
Added a new deviceIdentifier field (a UUID) to the tracing object. This value uniquely identifies multiple sessions from a given device (persisting until browser local storage is cleared). You may still use more specific identifiers, such as phone number or email, if you already know the user.
A new status code, VIDEO_PLAYS_INLINE_DISABLED, has been added. It’s returned when HTML5 videos default to the native full-screen player instead of playing inline, something that only occurs when the SDK is embedded in an iOS WebView with allowsInlineMediaPlayback set to false. This status code helps you identify and correct your WebView settings. For details on WebView integration, see .
Bug fixes
An issue has been resolved that was preventing the SDK from initializing properly when the DOM Storage API was disabled. This can occur on Android if the SDK is running inside a WebView without DOM storage explicitly enabled. For details on WebView integration, see .
Miscellaneous
Minor improvements.
Published on 2025-03-14
Implemented document auto-scanning with a streamlined process: the shutter button has been removed, and a full-screen camera view now automatically detects document types, performs quality checks, and ensures a seamless onboarding experience for end users with minimal friction. The SDK prioritizes selecting the optimal camera from the browser API's list but also allows users to manually switch cameras if focus issues arise.
The facial recognition module has been redesigned to deliver faster, more accurate face detection, providing a significantly smoother experience for the end user.
We have added a new operation error code, WASM_NOT_SUPPORTED. The SDK will terminate the session and trigger this error code if the browser lacks the essential features required to run a WASM module. However, this scenario is expected to be rare, as WASM is widely supported across modern browsers.
The frontImageId and backImageId fields in the scan object now reference the cropped document images. Additionally, two new fields, frontFrameImageId and backFrameImageId, have been introduced to store the complete frame images of the document. These full-frame images provide valuable context for troubleshooting and help fulfill regulatory requirements that necessitate documenting the scanning environment.
The base SDK size has been reduced to ~700 KB uncompressed, ~230 KB compressed, with the WASM module loaded via a CDN. Using a CDN offers several advantages:
Geographical proximity
Caching
Load distribution
The WASM module is available in two encodings: GZIP and Brotli. Brotli, which is widely supported by modern browsers, offers the best compression ratio. The current size of the WASM module is 3.7MB.
Enhanced Quality Check: Introduced a new verification step during document scanning to notify users if their fingers are obstructing the document.
Added support for landscape mode in the facial recognition step for tablets. This is to accommodate tablets that have the selfie camera in landscape configuration mode.
Expanded Document Support: Added compatibility for new document types, including SAU_ID (Premium Resident Version), SOM_ID (Somali ID), BHR_VL (Bahrain Vehicle License), JOR_VL (Jordan Vehicle License), and SAU_VL (Saudi Vehicle License).
Published on 2025-02-10
Reduced further the base size of the web SDK, from 1MB to 700KB uncompressed, 220KB compressed.
Added support for landscape mode in the facial recognition step for tablets. This is to accommodate tablets that have the selfie camera in landscape configuration mode.
Minor layout improvements. Note: This is a preview release, please dive into the details of the web SDK preview downloading the PDF below.
Published on 2025-01-17
Resolved an issue where no feedback message was displayed when users attempted to upload a file larger than 5MB.
Published on 2024-12-20
In this release we introduced auto scan similar to the mobile SDK where the SDK automatically detects and scans Id documents without requiring user interaction, making the process faster and more intuitive. Note: This is a preview release, please dive into the details of the web SDK preview downloading the PDF below.
Published on 2024-10-17
Enhanced SDK initialization to ensure that any resources from a previous SDK instance, such as the camera stream, are properly released. Additionally, a new method, dispose(), has been added to the UqudoSdk class, allowing you to forcibly release any resources related to the SDK.
Added support for a new document type: UAE_PASSPORT_DIGITAL, representing the digital version of the UAE passport. To accept this type, configure it as a separate document in your Enrollment Builder Configuration. For manual upload, we support the original 6-page PDF version and automatically extract the document from the third page.
Published on 2024-08-16
Introduced support for Qatar Vehicle License.
Enhanced the facial recognition step to provide a smoother experience for end users.
Fixed an issue where the selfie output in facial recognition appeared squeezed on some devices.
Published on 2024-07-12
Changed the image format to JPEG for the document photo-taking step.
Published on 2024-05-01
Introducing the new document type UAE_ID_DIGITAL aimed at clearly distinguishing between the digital and physical versions. If you want to accept the digital version, it is necessary to treat this as a separate document type in your configuration. It's important to note that starting from this version onwards, selecting the document type UAE_ID while the user provides the digital version will result in rejection. Likewise, selecting UAE_ID_DIGITAL while the user provides the physical version will yield the same outcome.
Published on 2024-04-10
Introducing a new configuration option for facial recognition, enabling 1:N face match verification. Once activated, following a successful facial recognition (confirming liveness and matching the face), the system initiates a search for the user's selfie within your tenant. If the selfie is not found, it is added, and the indexed facial features are stored in the database. The SDK result includes a unique ID in the face object, along with an indication of whether there was a match with a previously onboarded selfie. It's essential to store this unique ID in your system alongside the user's record, facilitating future searches for users with the same ID. Please be aware that this option requires a specific permission, otherwise, it will be disregarded. For further information or to explore this new feature, please reach out to your account manager.
Published on 2024-03-23
Optimized the screen resolution ratio for facial recognition across both mobile and desktop browsers. This adjustment aims to decrease the occurrences of rejections due to liveness check failures.
Detect unexpected errors occurring when the user's camera encounters an unknown issue. In such cases, we now promptly terminate the session and return the error code MEDIA_UNKNOWN_ERROR.
Published on 2023-12-15
Added an option to set the miniumum match level for facial recognition
Published on 2023-12-06
Corrected an issue where, on specific devices, the default selection of the rear camera led to capturing blurry images when taking a photo
Published on 2023-11-21
Added scan support for Iraqi ID
Published on 2023-09-23
We have introduced a new document type called MRZ. This document type allows you to scan documents with MRZ type TD1 or TD3
Resolved a bug that was impacting the facial recognition results on devices with zoom levels exceeding 100%
Published on 2023-08-18
In this latest release, we've revamped the facial recognition step. We've eliminated the nodding step, transforming facial recognition into a straightforward process easy as capturing a selfie. Under the hood, numerous enhancements have been implemented to ensure speed, simplicity, and enhanced accuracy. There are no changes in terms of business logic for your integration process. To accomplish the aforementioned outcome, we integrated AI capabilities directly into the Web SDK, resulting in a substantial increase in the SDK's size. To ensure the utmost efficiency with our latest SDK version, it's recommended to follow performance best practices. Enhance performance by reducing imports and incorporating techniques like lazy loading, code splitting, and memoization (specifically for React). To implement these strategies, consult the documentation of the framework you are utilizing.
Added support for Lebanon ID and Driving License
Published on 2023-05-13
Added new document type IND_PAN for Indian PAN card
Published on 2023-04-14
Added new document type PHL_DL for Philippines driving license
Added new document type BHR_DL for Bahrain driving license
Added new document type QAT_DL for Qatar driving license
Added new document type RSL_ID for Republic of Somaliland National ID Card
Published on 2023-03-17
Added new document type USA_DL for United States of America driving license
Published on 2023-03-03
Added new document type COD_VOTER_ID for Democratic Republic of the Congo driving license
Added new document type OMN_DL for Oman driving license
Added new document type OMN_VL for Oman vehicle license
Improved text definition for the upload and scan (taking picture) screen where now it is possible to define an array when for example you on-board two documents in the same session and you want to have a separate description for every document. The order is based on the order defined in the scan configuration for the enrollment flow.
From this version on in the SDK result all the fields containing a date will have an additional field with the same name plus the suffix "Formatted" that will contain the same date formatted in ISO-8601 format, e.g. '2011-12-03'
Published on 2023-02-17
Added new document type GBR_ID for UK resident id
Added new document type COD_DL for Democratic Republic of the Congo driving license
Document auto-scanning and facial recognition are powered by WASM technology. See some important details here
We have added new status codes for tracing in the document scanning and the facial recognition step. See the details here .
We've updated the names of some CSS classes. If you've overridden any CSS, please review the new list available here: .
Introduced support for the Analytics functionality. For more details, see .
Updated the facial recognition layout for desktop to enhance user experience and resolve the issue on Windows where the default zoom is set to 150%. We've added and updated some styles: for a complete list, please refer to the .
Enhanced the facial recognition layout for mobile and tablet. Fixed the oval's position, which sometimes misled users into placing their faces incorrectly. We've updated some styles: for a complete list, please refer to the .
Added virtual camera detection. If a virtual camera is detected, the SDK terminates the session with the error code MEDIA_VIRTUAL_CAMERA_DETECTED. For more details, see the .
Added Screen, Print and ID photo tampering detection to match the existing capabilities of the mobile SDK. Introduced a new verification object result within the SDK, including all details pertinent to the verification process. This feature supports your business logic by easily identifying potential issues. Refer to the for details.