> For the complete documentation index, see [llms.txt](https://support.hivo.co/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://support.hivo.co/changelog/1.6.7.md).

# 1.6.7

28 October 2026

## What's Changed

### \[New Feature] Minimum Face Sharpness for More Accurate Face Recognition

HIVO has introduced [**Minimum Face Sharpness**](https://support.hivo.co/client-profile/client-profile-settings/asset-and-metadata-management/recognition-settings#id-3.-configure-minimum-face-sharpness), a new setting that helps improve face recognition results by filtering out blurry faces, such as people in the background.

#### **🔹 What’s new?**

* **Customisable sharpness threshold:** Set a value from 0 to 100 to control how clear a face must be before it is included in Recognition.
* **Visual preview:** Use the example photo to see which faces meet your selected threshold and which will be ignored.
* **Improved recognition results:** Reduce unnecessary face detections in People and Approvals by focusing on clearer faces.

<figure><img src="https://2446612760-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F0sjg5GpIPr5HzBTJYqYe%2Fuploads%2Fg5Ey3FALvl1W9zGlfoJz%2FGroup%201168.png?alt=media&amp;token=de6918b9-799b-48fe-9b5b-e303a8adc86e" alt=""><figcaption></figcaption></figure>

#### **🔹 How to configure:**

Go to **Client Settings → Recognition Settings → Minimum Face Sharpness**, adjust the value using the slider or input field, and click **Save**.

⚠️ **User Right Required:** User must have the permission: **“Is Client Admin - User can manage the Client Profile”.**

***

### \[New Feature] Create Person – Automatically Tag Faces using Asset Metadata

HIVO has introduced **Create Person**, a new feature that automatically creates or matches People and tags faces using names stored in Asset Metadata, eliminating the need to tag each portrait manually.

#### **🔹 What’s new?**

* **Automatically create and match People:** Read names from the selected Metadata field (default: **Person**) and tag detected faces. Existing People are reused to avoid duplicates, even when names contain differences in capitalisation or spacing.
* **Process multiple Assets at once:** Create Person works with portraits containing exactly one detected, untagged face, making it easy to organise large collections of staff, player or team headshots.
* **Flexible workflows:** Run Create Person directly from **Library Actions**, **Asset Automations**, **Bulk Automations**, or enable it as an **Uploader Automation** for new Assets.
* **Supports manual confirmation:** When manual confirmation is enabled, face matches are sent to **Approvals > People > Suggested Matches** for review instead of being tagged immediately.
* **Clear processing results:** View how many People were created or matched, how many faces were tagged or suggested, and why certain Assets were skipped.

#### **🔹 Getting started**

* Go to **Client Settings > Recognition Settings** to configure the default **Create Person Name Field**.
* In the Library, select the portraits you want to process.
* Click **Actions > Create Person**, choose the Name Field, and click **Submit**.

👉 **Note:** Create Person only processes Assets with exactly one detected, untagged face and a valid name in the selected Metadata field. Existing face tags are never overwritten. For new uploads, enable both **Face Detection** and **Create Person** in Uploader Automations.

#### 👉 Learn more: **Create Person**

***

### \[New Feature] Start Combined Approvals Directly from Task Checklists

You can now start a **Combined Approval** directly from an **Asset Approval** checklist item, allowing you to submit multiple assets for approval in a single request.

#### **🔹 What’s new?**

* Simply select **2–100 eligible assets** from the checklist and click **Start Approval**.
* Once all assets in the checklist item are approved, the item is automatically marked as completed. If an asset's approval status changes and the completion conditions are no longer met, the checklist item is automatically reopened.

<figure><img src="https://2446612760-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F0sjg5GpIPr5HzBTJYqYe%2Fuploads%2FL3jI96rs8lppBBnz19Ww%2FGroup%201101%20(2).png?alt=media&amp;token=9ebc442c-0e00-4307-943f-d3f04da045c6" alt=""><figcaption></figcaption></figure>

This improvement simplifies approval management, reduces repetitive steps, and helps keep task progress aligned with asset approval statuses.

#### 👉 Learn more: [**Task Checklist**](https://support.hivo.co/tasks/task-detail-page#task-checklist)

***

### \[New Feature] Approval Categories

You can now organise and manage approval requests using **Approval Categories**.

Assign one or more predefined categories to **Asset Approvals** and **Combined Approvals**, making it easier to find, filter, track, and report on approvals.

<figure><img src="https://2446612760-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F0sjg5GpIPr5HzBTJYqYe%2Fuploads%2FLRqaxrQHccEYh1D0qTDE%2FGroup%201161.png?alt=media&amp;token=c968438e-6ee9-4846-9a7c-bc1c03c9e3f6" alt=""><figcaption></figcaption></figure>

#### **🔹 What’s new?**

* **Custom Categories:** Admins can configure up to 64 categories in **Client Profile → Settings → Approval Settings → Asset Approval:** [**Configure Approval Category Settings**](https://support.hivo.co/client-profile/client-profile-settings/workflow-and-automation/approval-settings#id-4.-configure-approval-category-settings)**.**
* **Optional or Required:** Admins can choose whether users must select at least one category when creating or editing an approval request.
* **Improved Visibility:** View assigned categories in Approval Tables, Approval Details, and Kanban View.
* **Filter & Export:** Filter approvals by category and include category information in CSV exports.

<figure><img src="https://2446612760-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F0sjg5GpIPr5HzBTJYqYe%2Fuploads%2FJ9oNry5zfQaiIXFb8rvC%2FGroup%201156.png?alt=media&amp;token=7b93af65-d925-4317-bed5-f83d1632b81a" alt=""><figcaption></figcaption></figure>

👉 **Note:**

* Approval Categories are available for Asset Approvals and Combined Approvals only, not Template Approvals.
* Disabling Approval Categories or removing category options does not affect categories already assigned to existing approvals.

#### 👉 Learn more: [Adding Approval Categories to an approval request](https://support.hivo.co/approvals/asset-approvals#id-4.-adding-approval-categories-to-an-approval-request)

***

### \[New Feature] AI Auto Tags – Smarter, More Relevant Image Tagging

We've introduced **AI Auto Tags**, an AI-powered feature that suggests more accurate and relevant tags for image Assets by understanding the overall content and context of each image.

{% hint style="info" %}
AI Auto Tags is an Additional Feature that is enabled per account by a HIVO Admin. If you would like it turned on for your organisation, please contact the HIVO Support team.
{% endhint %}

* Previewing Automatic Tag feature with AI Auto-Tag"

<figure><img src="https://2446612760-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F0sjg5GpIPr5HzBTJYqYe%2Fuploads%2FiLMK3nfXz0GNqxykdDpM%2FGroup%201153.png?alt=media&amp;token=6bc91267-1781-476b-8088-4741c7c3e498" alt=""><figcaption></figcaption></figure>

#### **🔹 What’s new?**

* **Smarter tag suggestions:** AI identifies more specific details in images, improving tag relevance and making Assets easier to find.
* **Works across HIVO:** Available in Tag Prediction during upload, Suggest Tags on Asset Detail, and bulk tagging.
* **Existing tag controls remain:** Minimum Confidence and Blacklisted Tags settings still apply. Suggested tags are saved as Automatic Tags and can be reviewed or removed as usual.
* **Customisable AI context:** Client Admins can configure **AI Auto Tags Context** in Tag Settings to help AI suggest tags relevant to their organisation. AI only tags what is visible in the image.

<figure><img src="https://2446612760-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F0sjg5GpIPr5HzBTJYqYe%2Fuploads%2FxJxo490IovJtGroD8ejo%2FGroup%201164.png?alt=media&amp;token=82ff5fdb-ff89-4788-ac37-f98f5ce9381e" alt=""><figcaption></figcaption></figure>

👉  **Note:**

* AI Auto Tags applies to images only. Videos continue to use Amazon Rekognition.
* If AI tagging fails, HIVO automatically falls back to Amazon Rekognition.
* AI tagging may take longer, especially when processing multiple images.
* Changes to AI Auto Tags Context only affect future tagging, not existing tags.

**Availability:** AI Auto Tags is an Additional Feature that must be enabled by a HIVO Admin. Contact HIVO Support to activate it for your organisation.

#### 👉 Learn more: [**AI Auto Tags**](https://support.hivo.co/assets/tagging-assets#ai-auto-tags-images)

***

### \[Improvement] Regenerate Web and Thumbnail from Any Asset Version

You can now regenerate an Asset’s **Web** and **Thumbnail** versions from any eligible version, such as **Approved** or **V2**, instead of only the **Original**. This also updates the Asset’s preview in the Library to reflect the selected version.

#### **🔹 What’s new?**

* **Upload New Version:** Choose to regenerate Web and Thumbnail when uploading a new version.
* **Asset Automations:** Regenerate existing Web and Thumbnail versions using the selected Asset version.
* **Image Editor:** When saving edits as a new version, choose whether to regenerate Web and Thumbnail so the Library preview reflects your changes.

👉 Follow the step-by-step guide in the demo here:

{% @howdygo/embed url="<https://app.howdygo.com/share/6eba9931-d7c7-4ad8-a71e-a5240ecbe2f0>" %}

#### **🔹 How to enable**

* Go to **Client Profile → Settings → Version Settings → Web and Thumbnail**&#x20;
* And enable **Allow regenerating Web and Thumbnail from any version**.

<figure><img src="https://2446612760-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F0sjg5GpIPr5HzBTJYqYe%2Fuploads%2FZMTQiejaPTgw3SH0IF3L%2FGroup%201124.png?alt=media&amp;token=d55ad420-fe1f-4ce9-99b6-4547269c8c10" alt=""><figcaption></figcaption></figure>

**Note:** This setting is disabled by default and can only be enabled by Client Admins. Regeneration is not available for Web, Thumbnail, or versions hidden by default. Bulk Automations in the Library remain unchanged.

#### 👉 Learn more[ **Regenerate Web and Thumbnail Versions**](https://support.hivo.co/assets/asset-versions#id-3.-regenerate-web-and-thumbnail-versions)


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://support.hivo.co/changelog/1.6.7.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
