There’s a particular kind of friction that accumulates quietly in day-to-day work with 3D sensors – not the dramatic kind that blocks progress, but the small, repetitive kind that costs time and introduces errors without ever announcing itself.

With PhoXi Control 1.17, Photoneo has replaced that workflow entirely. The result is something that integrators and engineers who’ve spent time in 3D software will recognise immediately – and those who haven’t will find intuitive within minutes.

What a Region of Interest Actually Does

Before getting into the interface change, it’s worth being precise about what ROI configuration accomplishes and what it doesn’t – because the distinction matters for understanding where this update delivers its real value.

A Region of Interest in PhoXi Control defines a spatial volume within the sensor’s field of view. Point cloud data captured outside that volume is excluded from further processing. Everything the sensor sees, it still scans – the ROI doesn’t alter the acquisition process, the exposure, or the depth calculation. What it does is define a boundary after capture: only geometry inside the box gets passed downstream.

This has two immediate practical effects. 

1. It removes irrelevant structure from your data – conveyor surfaces, fixture edges, background walls, anything that’s in the sensor’s sight line but not relevant to the task. Cleaner data means faster and more reliable downstream processing, whether that’s a localization algorithm, a grasp planner, or an inspection routine. 
2. If you’re consuming the point cloud in a downstream application – a Python script, a robotics pipeline, a custom integration – your code is only processing the subset of data that actually matters. The computational load shrinks in proportion to how aggressively you’ve cropped your volume.

To be clear: scan time doesn’t change. Transfer time doesn’t change. What changes is everything that happens after the data lands in your pipeline.

The Old Approach: Numeric Input

Previously, setting up a 3D ROI in PhoXi Control meant entering numerical values for each axis – X minimum, X maximum, Y minimum, Y maximum, Z minimum, Z maximum. Six fields defining a box in space, filled in by a person who either measured their scene carefully, iterated through trial and error, or copied values from a previous setup.

This works. It’s unambiguous – coordinates are coordinates – and for teams with established workflows and known scene geometries, it was serviceable. But it has a limitation: there’s no intuitive spatial feedback. 

You’re describing a three-dimensional volume through a one-dimensional interface, and the only way to verify you got it right is to scan and inspect the result. If your box was slightly off – too narrow, clipping part of the object of interest, or leaving too much surrounding noise in – you adjusted the numbers and scanned again.

For initial setup, thisI’m  iteration cycle adds up. For anyone setting up a new cell, a new part type, or troubleshooting an edge case in the field, it was friction that didn’t need to exist.

The New Approach: An Interactive Bounding Box

PhoXi Control 1.17 replaces this with a fully interactive 3D bounding box, adjustable directly within the 3D viewer.

The concept will be immediately familiar to anyone who has used point cloud software – CloudCompare being the most widely known example in industrial and research contexts – or any modern 3D modelling environment. A visible, manipulable box is overlaid on your scan. You grab its faces, edges, or corners and drag them to the boundaries you want. The volume updates in real time, against the actual geometry of your scene.

The practical difference is significant. Instead of estimating coordinates and verifying them through repeated scans, you’re working directly with the spatial reality of your setup. The object you’re trying to isolate is visible in the viewer. The box you’re drawing around it is visible in the viewer. The relationship between the two is immediate and legible.

You can move the entire bounding box to reposition it within the scene. You can extend or contract it along any axis independently. For setups where the ROI needs to sit at an angle, or where the relevant volume is asymmetric, you’re adjusting handles rather than solving geometry mentally and translating it into six numbers.

Where This Matters in Practice

The applications where this update removes the most friction tend to share a common characteristic: the scene geometry is known but not perfectly regular, and the ROI needs to be fitted around something real rather than specified from a drawing.

• New cell commissioning. When setting up a new application – new part type, new fixture, new mounting position – the first pass at ROI configuration is almost always iterative. The interactive bounding box compresses those iterations substantially. You see your scene, you fit your box, you’re done.
• Field adjustment and maintenance. If a sensor has been repositioned, or a fixture has shifted, ROI values that were correct are now wrong. Correcting them via numeric input requires re-measuring or re-iterating. Correcting them via the interactive viewer requires dragging a face to the right place.
• Partial scene processing. Many real-world setups only care about a subset of what the sensor sees – a bin rather than the full conveyor, a workpiece rather than the surrounding fixture, a pallet position rather than the full warehouse floor. Precisely cropping to exactly the relevant volume is easier when you can see what you’re cropping to.
• Communicating setups across teams. A bounding box visible in the viewer is a self-documenting configuration. A set of six coordinate values requires context to interpret. For teams where sensor configuration is shared between integration engineers and application developers, this is a non-trivial quality-of-life improvement.

A Small Change That Respects Your Time

The 3D ROI update in PhoXi Control 1.17 is not a fundamental capability change. You could always define a Region of Interest. What changes is how much effort that takes and how accurately you can do it without specialized geometric knowledge or iterative trial and error.

That might sound like a modest improvement. In practice, any recurring workflow that becomes meaningfully faster and less error-prone compounds across every setup, every commissioning task, every field visit. The engineers closest to these workflows will feel the difference immediately. The ones further from them will notice it in the shape of faster deployments and fewer support interactions about ROI configuration edge cases.

Available Now 

PhoXi Control & Firmware 1.17 is available for download. Updated documentation, API references, and working code examples for all supported languages are published at github.com/photoneo-3d, documentation can be found here – docs.photoneo.com.

Check out the Knowledge Base here.