Recent IOS development of our web viewer for higher performance on ios devices.
IOS 27 3DGS viewer
IOS 27 3DGS viewer
We’ve recently developed a real-time collaboration feature for 3D experiences created on top of Playcanvas engine.
With one click, a user can create a room, generate a link, and invite others to join the same session directly through their browser—without installing additional software or going through a complicated setup process.
The room host can guide everyone through the environment, while participants in different locations see the same camera movement in real time.
Participants are also not permanently locked to the host’s view. At any time, they can leave the guided camera and freely explore the environment themselves. Other members can see each participant’s position within the 3D space, making it much easier to discuss specific areas and understand exactly where everyone is referring to.
Potential applications include:
This system is not limited to one specific type of 3D content. It can potentially be applied to point clouds, 3D Gaussian Splatting, traditional 3D models and other spatial datasets.
Our broader goal is to help organisations get more value from the 3D data they already have. Instead of leaving valuable datasets inside specialist software or treating them as static deliverables, we want to turn them into accessible environments for exploration, communication and collaboration.
Our next development stage will focus on:
As long as the information can be represented in 3D, we want to explore how it can be made more accessible, interactive and useful.
We’d be interested to hear how people working with point clouds, 3DGS, digital twins, architecture, surveying, heritage or remote site reviews might use this.
You can now look around the scene by moving your phone, which feels much more natural on mobile.
Still testing it, but keen to hear what you think.
https://reddit.com/link/1usjq2x/video/72oixd68qdch1/player
We recently scanned this real-world environment and brought the point cloud into our browser-based 3D viewer.
Instead of only rotating around the model, users can move through the space more like a video game—walking around, exploring different areas, and viewing the scan from a first-person perspective.
We are currently testing ways to make large point cloud datasets easier to navigate and more interactive, including collision, clickable elements, measurements, annotations, and access to spatial information.
Here is a short recording of the experience. I’d be interested to hear what people think about using point clouds in this way, and what other tools or features would be useful.
Built by Glim Sphere 3D using our in-house developed viewer.
#pointcloud #3DScanning #LiDAR #WebGL #DigitalTwin #RealityCapture
Been messing around with point cloud viewers for a while and got kind of tired of the usual options — either a desktop app like CloudCompare that needs some setup and a decent GPU, or a web viewer where you're just orbiting around a static blob and have to export somewhere else if you want to measure anything.
So we tried something different: you can actually walk through the point cloud in first person, like you're standing in the space, and just measure stuff directly while you're in there. Runs in a browser, no install, and it's held up fine on my phone and an older laptop too, not just a proper workstation.
Attached a video below of it in action, probably easier to see than to explain.
The project in the video was captured as 3DGS, but I tested the same walkthrough/measuring on some older "regular" point cloud data too and it works basically the same way — which was kind of a nice surprise, since we've got a bunch of old scan data sitting around that nobody really opens anymore.
Curious if others here have run into the same annoyances with viewing/delivering point cloud data.
Glim Sphere 3D
We recently scanned a heritage building and tested a custom shader inside our 3D viewer to highlight specific areas of the site.
The idea was to make important parts of the building easier to read in the viewer, rather than relying only on text labels or a standard 3D scan. By highlighting selected areas directly in the scene, users can more quickly understand where to look and what parts of the building are significant.
It’s still an early test, but I think this kind of visual guidance could be useful for digital heritage, museums, architectural documentation, and interactive site interpretation.
Glim Sphere 3D
Hi everyone,
We’ve been working on a browser-based 3D viewer, and I wanted to share a small UI component we recently added: a Section Tool.
It allows users to cut through a 3D scene with an interactive section plane, making it easier to inspect interiors, spatial relationships, and hidden parts of a reconstructed environment.
This is still a work-in-progress feature, but the goal is to make scanned or reconstructed 3D spaces easier to explore without needing CAD/BIM software.
I’ve included a short screen recording of the tool in use.
Would appreciate any feedback on the UI, interaction flow, and whether this feels useful for 3D scans, architecture, heritage documentation, or Gaussian Splatting / spatial capture workflows.
Glim Sphere 3D
Hi everyone,
We recently scanned a local restaurant using the XGRIDS K2 and processed the data into a 3D Gaussian Splatting scene. We then brought the reconstructed scene into our own web-based 3D viewer built based on Playcanvas Engine and recorded a short walkthrough/demo.
The workflow was roughly:
We are still improving the viewer and overall workflow, so I’d be interested to hear what people think about the visual quality, loading experience, navigation, and possible use cases.
Would appreciate any feedback or suggestions.
https://glimsphere3d.com.au/Glim_vtour_LILHKN/index.html
Here is a link if you want to have a look.