3D Geodata Academy

Limited time
Free Innovation Sprint: See Production Systems Live
Join Newsletter
Login
3D Geodata Academy logo, back to homepage
Entreprise
Start Free

3D Intelligence Reports

Daily · Verified · Free

3D Intelligence Reports

Daily verified intel on 3D and spatial AI. Five finds a day, checked the same day they ship, with my take on why each one matters.

Latest report: June 22, 2026 · all links verified that day

How this gets made

Every morning I go hunting through arxiv, funding portals, job boards and company news. A lot of it doesn’t survive verification. What survives ends up here: one paper with working code, one standout job with real salary data, four funding doors with live deadlines, one tool worth your weekend, and one market move that tells you where this field is heading.

Why bother? You shouldn’t need three hours a day to stay ahead of 3D and spatial AI. I check every link the same day it goes out, my take sits next to every pick, and the archive below grows into a searchable library of the field’s trajectory, one verified day at a time.

Today’s report

June 22, 2026

**Theme of the day:** 3D perception is leaving the data center for the thing that moves. A drone that reconstructs as it flies, a robot that fuses its sensors into a scene, glasses that anchor to the world on-device. The plumbing to ship splats and a funding door for novel ideas round it out.

Scroll or use the arrows · hover a visual to play it

Paper published 2026-06-16

MoonSplat: Monocular Online Gaussian Splatting with Sim(3) Global Optimization

arxiv 2606.17935 (SIGGRAPH 2026) · arxiv 2606.17935

Guo Pu, Yixuan Han, Haofeng Li, Yao Zhang, Hui Zhou, Zhouhui Lian (Peking University, per the authors' prior work)

Live 3D from one moving camera

MoonSplat turns a single moving camera, including a drone feed, into a live photorealistic Gaussian splat map while it tracks where the camera is, the classic SLAM-meets-splatting problem. Its move is a Sim(3) global optimization layer that jointly corrects scale, rotation, and translation across the whole trajectory, the two things online splat methods break on: fragile pose estimation and collapse over long or large-scale sequences. The team built a real drone-based capture rig on top of it, which is exactly the fly-and-reconstruct workflow the geospatial crowd runs. Code, run scripts, config, and demo data are public, and it is positioned as SIGGRAPH 2026 work, a strong niche plus hype combination. Pretrained weights and the camera-intrinsics estimation module are not released yet.

Florent’s take

Monocular SLAM has always drifted. You walk or fly a long path, small errors in scale and pose pile up, and by the time you close a loop the map no longer lines up with itself. MoonSplat's bet is to treat the whole trajectory as one thing and correct it with a Sim(3) global step, scale and rotation and translation together, instead of patching frame by frame. That is the right place to fight drift. The honest catch: the intrinsics module is not out yet, so today you feed it your own camera calibration, and there are no pretrained weights, it is a research drop, not a turnkey tool. If you test one thing, fly a real outdoor pass of a few hundred frames and check whether the scale holds when the loop closes, because that is where this lives or dies.

Read the paper Code on GitHub
Read the full breakdown →
Job verified open 2026-06-22

Computer Vision AI & ML Engineer

Skild AI · San Mateo, CA (on-site)

$180K-$300K + equity (est.)

Florent’s take

Read the requirements, not the title. A robot foundation-model company is asking for object detection, segmentation, and tracking, but also 3D scene understanding with multi-sensor fusion across RGB-D, LiDAR, and stereo, with 3D geometry named explicitly. That is the spatial-data skillset this audience already has, point clouds, sensor fusion, geometry, now sitting at the center of embodied AI. The robot's perception layer is classic 3D and geospatial computer vision wearing a new badge. Notice too that half the job is data engineering: pipelines, evaluation, annotation strategy, which says owning the 3D data path is the durable moat, not just the model. If you go for it, lead with an end-to-end 3D perception pipeline you deployed on real hardware, not a clean benchmark number.

View the live posting
Read the full breakdown →
Opportunity deadline checked on the call page

ESA Open Space Innovation Platform, Open Discovery Ideas Channel

Europe (+ Global) · deadline Jul 17, 2026 · 25 days left

Studies EUR 20K to 100K (up to 100%), Early Tech Dev EUR 90K to 175K (up to 100%), Co-sponsored Research up to EUR 90K (50%). Entry is a one to two paragraph idea.

Florent’s take

This is the rare funding door built for exactly this audience. A solo researcher or a two or three person spatial-AI company can enter with a one or two paragraph idea, no consortium, no co-applicants, no 45 page proposal. Selection rides on novelty, not track record, so a sharp 3D or geospatial-AI concept can win a fully funded study: neural point-cloud compression for satellite downlink, on-board 3D scene understanding, splat-based planetary mapping. It is continuously open, evaluated the third Friday of each month, so there is no missed-the-call, the next gate is July 17. The mistakes people make: over-engineering it into a Horizon-style document, pitching something incremental, or submitting and going quiet, you have to mark the idea ready for evaluation or it never gets seen. This week: make the account, read three or four already-funded ideas to calibrate the novelty bar, then write the tight pitch.

Open the call page
Read the full breakdown →
Tool released 2026-06-16

SplatTransform v2.6.0

One CLI for every splat format

SplatTransform is PlayCanvas's open-source (MIT) command-line tool for Gaussian splats: it reads PLY, Compressed PLY, SOG, SPZ, SPLAT, KSPLAT, LCC and LCC2, and writes PLY, Compressed PLY, SOG, SPZ, GLB, CSV, an HTML viewer, LOD, voxel, and WebP. v2.6.0 adds read support for LCC2 (the newer XGRIDS scanner format) and halves its peak read memory, parallelizes SOG writing across a worker pool, hardens PLY header parsing, and fixes a black-render regression after the engine 2.19 upgrade. Install with npm (npm install -g @playcanvas/splat-transform); code and full notes are on GitHub.

Florent’s take

This is the unglamorous tool that actually unblocks a splat pipeline. You get a .ply out of your trainer, and then you hit the wall: getting it into a web viewer, a game engine, or a compressed delivery format. SplatTransform is the one command-line tool that converts between the formats, transforms the geometry (translate, rotate, scale), filters by box or sphere, and builds LOD and voxel trees, so it drops straight into automation. The honest limit: it is a converter and editor, not a trainer, it will not make your reconstruction better, it just moves and reshapes what you already have. The new LCC2 support matters because XGRIDS scanner output keeps showing up in survey and reality-capture work, and this gets it into open formats. Try it on one heavy splat: convert a .ply to .sog and emit a one-line HTML viewer, no server needed.

Repository on GitHub
Read the full breakdown →
Market announced 2026-06-16

Snap unveiled SPECS, its first fully standalone consumer AR glasses (no tether, no puck), with pre-orders at $2,195 and shipping in fall 2026.

Snap

Snap is the first major player to put a genuinely standalone consumer AR headset in front of real buyers, not just developers. The hardware runs two Snapdragon processors, one of them dedicated to computer vision, precisely because anchoring digital content to the real world is a live 3D spatial-mapping and localization problem, the exact stack this field works in. SPECS turns spatial intelligence from a back-end pipeline into a mass-market wearable surface, which pulls real-time SLAM, scene understanding, and splat-grade capture toward the consumer edge. It also reframes world-model work as something that has to run on-device at 7 millisecond latency, not in a data center.

Florent’s take

The deeper signal is two threads this audience already follows starting to meet: world models in the cloud, built at simulation scale, and spatial perception on-device, judged by latency. Snap is betting the prize is at the edge, a dedicated vision chip and a 7 millisecond budget say real-world anchoring has to be local. Worth weighing against the hype: at $2,195 with a 51 degree field of view this is still prosumer, and it lands in the same stretch as Apple putting native splats in RealityKit, so the glasses-as-spatial-platform race is openly contested now. The thread for anyone doing 3D: the capture and reconstruction skills you build today are what fills these glasses tomorrow, running on the device, not on a server. Watch whether the developer tools start expecting your splats as input.

Read the source
Read the full breakdown →
Open today’s report at its own URL

The intel library

One card per daily report, newest first. Pick a track, search the items, or keep only the funding calls still open.

3D Intelligence Report – June 22, 2026

Open report ↗
  • Paper MoonSplat: Monocular Online Gaussian Splatting with Sim(3) Global Optimization
  • Job Computer Vision AI & ML Engineer, Skild AI
  • Funding ESA Open Space Innovation Platform, Open Discovery Ideas Channel (25 days left)
  • Tool SplatTransform v2.6.0
  • Market Snap: Snap unveiled SPECS, its first fully standalone consumer AR glasses (no tether, no puck), with pre-orders at $2,195 and shipping in fall 2026.

3D Intelligence Report – June 16, 2026

Open report ↗
  • Paper BRDFusion: Physics Meets Generation for Urban Scene Inverse Rendering
  • Job Senior Computer Vision Engineer (Localization), Niantic Spatial
  • Funding Marie Sklodowska-Curie Postdoctoral Fellowships 2026 (85 days left)
  • Tool ComfyUI v0.23.0
  • Market Cesium: Cesium ion reconstruction now runs point cloud and Gaussian splat jobs with no mesh step

3D Intelligence Report – June 15, 2026

Open report ↗
  • Paper Wild3R: Feed-Forward 3D Gaussian Splatting from Unconstrained Sparse Photo Collection
  • Job Research Engineer / Scientist (SLAM), World Labs
  • Funding 4 funding doors, nearest deadline in 15 days
  • Tool NVIDIA 3D Object Reconstruction v0.2.0 (Apache 2.0)
  • Market NVIDIA: Physical AI agent skills: agents that automate the physical-AI workflow, including turning fleet-captured video into editable 3D scenes for simulation and synthetic data, built on Cosmos 3 with Isaac Sim and Omniverse

3D Intelligence Report – June 12, 2026

Open report ↗
  • Paper Light-WAM: Efficient World Action Models with State-Fusion Action Decoding
  • Job Senior System Software Engineer – Neural Graphics Performance, NVIDIA
  • Funding 4 funding doors, nearest deadline in 6 days
  • Tool REFINE initial release
  • Market Leica Geosystems (Hexagon): Leica RTC series launch: RTC300/500/700 terrestrial laser scanners with real-time field-to-office streaming

3D Intelligence Report – June 11, 2026

Open report ↗
  • Paper Envision4D: Envisioning Visual Futures via Feed-forward 4D Gaussian Splatting for Autonomous Driving
  • Job Research Engineer / Scientist (3D Reconstruction), World Labs
  • Funding 4 funding doors, nearest deadline in 24 days
  • Tool PyTorch Geometric v2.8.0
  • Market Apple: RealityKit ships native 3D Gaussian splat rendering at WWDC26; Apple Maps Flyover moves to splats

3D Intelligence Report – June 10, 2026

Open report ↗
  • Paper WorldOlympiad: Can Your World Model Survive a Triathlon?
  • Job Technical Lead, Computer Vision, Niantic Spatial
  • Funding 4 funding doors, nearest deadline in 28 days
  • Tool SuperSplat v2.27.4
  • Market World Labs: A Functional Taxonomy of World Models: renderers, simulators, planners

Or browse by day: June 22, 2026 · June 16, 2026 · June 15, 2026 · June 12, 2026 · June 11, 2026 · June 10, 2026

Questions, answered

What is a 3D Intelligence Report?+

One day of verified intel on 3D and spatial AI, in five tracks: a paper with working code, an open job with disclosed pay, four funding doors across four regions, a tool release worth your upgrade time, and a market move that changes the field. I add my take on each one so you know why it matters, not just that it happened.

How do you verify the items?+

I fetch every link the same day I publish. A paper needs real code in the repo, a job needs a live posting, a deadline needs the official call page, a release needs the changelog. If a source doesn't hold up, I drop it. Nothing gets padded and nothing gets invented.

How often does a new report come out?+

One report per day. Each one lives at its own URL, and this page always shows the latest in full, with the full intel library right below.

How do I find older items?+

The intel library on this page lists every item from every report. Filter it by track, search by title, company or region, or flip the deadlines toggle to see only the funding calls that are still open. Each item links to its day's report and to the original source.

How do I get it by email?+

Drop your email in the form above and tick the consent box. You'll get the daily report in your inbox, and you can unsubscribe with one click at any time.

Don't want to check back every day?

Five verified finds, my take on each, one short email a day.

Scroll to Top