---
title: "Enhanced bounding box visualizations"
slug: "relnote-12-enhanced-bounding-box-visualizations"
updated: 2025-05-16T11:40:19Z
published: 2025-05-16T11:40:19Z
canonical: "docs.akridata.ai/relnote-12-enhanced-bounding-box-visualizations"
---

> ## Documentation Index
> Fetch the complete documentation index at: https://docs.akridata.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Enhanced bounding box visualizations

The [Virtual Column](https://docs.akridata.ai/docs/relnote-12-virtual-catalog-columns) feature forms the building blocks for richer visualization functionalities introduced in this release.

## 3D bounding box visualization

![](https://cdn.document360.io/3e9d4528-fbc6-4948-a804-8ee7068e7ac3/Images/Documentation/Screenshot%20from%202023-02-03%2019-17-02.png)

As seen in the above image, this release brings in the capability to project 3D bounding boxes on thumbnails and full-resolution images. The 3D bounding boxes are specified by adding virtual columns to the view. The below shows one of the 3D bounding box type definitions that is defined using 8 corners and image sizes.

![](https://cdn.document360.io/3e9d4528-fbc6-4948-a804-8ee7068e7ac3/Images/Documentation/Screenshot%20from%202023-02-03%2019-19-09.png)

Another 3D bounding box definition that is provided uses the (camera's intrinsic and extrinsic parameters), rotation, translation etc as shown below.

![](https://cdn.document360.io/3e9d4528-fbc6-4948-a804-8ee7068e7ac3/Images/Documentation/Screenshot%20from%202023-02-03%2019-22-19.png)

## Bounding box visualization in all types of jobs

In the previous release, 2D prediction and ground truth bounding boxes were available only in an **'analyze object detection'** type of job. In this release, 2D prediction and ground truth bounding boxes are visualizable in all types of jobs.