Erik Christensen

Try out the tool here! scanspace.nz/pages/scale

It only took 4 years to get this out, but its here. A video on how to design, build and use Scale bars in your photogrammetry! Please share it around! youtube.com/watch?v=mMyV_F…

Holy shit... Stanford just proved that GPT-5, Gemini, and Claude can't actually see. They removed every image from 6 major vision benchmarks. The models still scored 70-80% accuracy. They were never looking at your photos. Your scans. Your X-rays. Here's what's really going on: ↓ The paper is called MIRAGE. Co-authored by Fei-Fei Li. They tested GPT-5.1, Gemini-3-Pro, Claude Opus 4.5, and Gemini-2.5-Pro across 6 benchmarks -- medical and general. Then silently removed every image. No warning. No prompt change. The models didn't even notice. They kept describing images in detail. Diagnosing conditions. Writing full reasoning traces. From images that were never there. Stanford calls it the "mirage effect." Not hallucination. Something worse. Hallucination = making up wrong details about a real input. Mirage = constructing an entire fake reality and reasoning from it confidently. The models built imaginary X-rays, described fake nodules, and diagnosed conditions -- all from text patterns alone. But that's not the scary part. They trained a "super-guesser" -- a tiny 3B parameter text-only model. Zero vision capability. Fine-tuned it on the largest chest X-ray benchmark (696,000 questions). Images removed. It beat GPT-5. It beat Gemini. It beat Claude. It beat actual radiologists. Ranked #1 on the held-out test set. Without ever seeing a single X-ray. The reasoning traces? Indistinguishable from real visual analysis. Now here's what should terrify you: When the models fake-see medical images, their mirage diagnoses are heavily biased toward the most dangerous conditions. STEMI. Melanoma. Carcinoma. Life-threatening diagnoses -- from images that don't exist. 230 million people ask health questions on ChatGPT every day. They also found something wild: → Tell a model "there's no image, just guess" -- performance drops → Silently remove the image and let it assume it's there -- performance stays high The model enters "mirage mode." It doesn't know it can't see. And it performs BETTER when it doesn't know it's blind. When Stanford applied their cleanup method (B-Clean) to existing benchmarks, it removed 74-77% of all questions. Three-quarters of "vision" benchmarks don't test vision. Every leaderboard. Every "multimodal breakthrough." Every benchmark score you've seen this year. Built on mirages. Code is open-sourced. Paper is live on arXiv. If you're building anything with multimodal AI -- especially in healthcare -- read this paper before you ship. (Link in the comments)

@80Level I had the joy of leading the 3D scanning for the Minecraft movie, and the scan of Malgoshas cloak was something else entirely. really proud of what we managed to do. The CG cloak matched the reference cloak so well.

Wētā FX shared the visual effects work done for A Minecraft Movie and recalled how Malgosha was brought to life. Listen to the befores & afters podcast: 80.lv/articles/see-h…

@jo_chemla sounds good. I'll add it to my todo list.

@shrogg ArUco markers are useful for perspective correction via custom OpenCV pipelines, when taking batches of single photo of flat 2D items, undistorted using the lens/calibration profile, and then transformed to a scaled, orthographic coordinate system. See docs.opencv.org/4.x/d5/dae/tut…

I've just released a Free #Photogrammetry Scale bar design tool! scanspace.nz/pages/Scale This tool is designed to make the creation of scale bars at home really easy, and accurate enough for most tasks (as long as you print at 100% scale!) Let me know what you think!

@allabout3dblog I'm yet to properly test as I don't have metashape Pro, but I believe Metashape might support April Tags? The April Tag 16H5 and 36H11 sets seem to be most universal. The Circular Tags are not cross-compatible between RS/Metashape/Zephyr

@shrogg Amazing! Which scale types are compatible with both, Metashape and RealityScan?

• Elon Musk changed Iran's flag on X. • Removed verification from Iranian leaders accounts. • Platform barely shows news from Iran. • Boosting only Israeli and American propaganda. And they still call this "free speech."

@jo_chemla What software uses these custom markers? I can absolutely look at implementing other marker families if they have uses. Background checkerboard grid is a great idea, I'm considering adding other noise profiles, as well as custom image backgrounds.

@shrogg Nice tool, thanks for sharing that to the community! Would be great to add custom markers like aruco Might also make sense, the way you added noise background, to add background checkerboard grid (lines rather than black/white cells), rather than as a custom shape

Fix is live!

Theres a bug currently with Reality Scan Tag generation, where the dictionary it uses isn't able to pull any tags generated by Reality Scan, and instead is using hte fallback April Tag generator, creating tags that have incorrect IDs and structure for Reality Scan. Fix coming!

@beforesmag @Framestore Oh how cool! On the first and Second seasons of Prehistoric Planet I 3D printed heads/feet of our Dinos to use as Lighting/footprint reference, but nothing to this level as we were heavily limited in resources. We ended up using the Foot quite a bit for the nest sequence in S1.

I am obsessed with this new featurette released by Apple on the puppetry in 'Prehistoric Planet: Ice Age'. Watch in full at @beforesmag, and stay tuned for an in-depth podcast about @Framestore's work on the series with VFX supervisor Russell Dodgson. beforesandafters.com/2025/12/04/thi…

Well if this isn't the best news. Get Fucked guppy and your goons sleucing money away from the city.

Hey everyone! I am running a survey on the #photogrammetry industry and would love for people of all skill levels who use photogrammetry to fill it out. Please share this around! Ill release an analysis of the data in a few months! forms.gle/ktJHiJb2DEeU1i…

Just updated the repo for Image Space, my #Photogrammetry and #3DGS image processing toolkit. Added Dark mode and heaps of small fixes. github.com/ErikScansNz/sc…

I've just launched the latest update to my free #Photogrammetry Image processing software, Image Space! Check it out! :D scanspace.nz/blogs/news/ima…

Ooooh this is it lads. Added some network processing to my image processor. #Photogrammetry #3dgs

I'm hopefully pushing out a very massive update to the #Photogrammetry Image Processor today. new features: Smarter importing of images - Import Images as Datasets Customized export syntax - Export using a wide range of controls! Network Processing - Process LOTS all at once!

I have just released the a first version of my Free Photogrammetry Image Processing tool! scanspace.nz/blogs/news/bat… Features: Automatic Chart Detection Manual Chart Selection & Flatten Batch Colour Correction Exposure Normalization #Photogrammetry #3DScanning #3DGS #Software

The shared values