Image created with gemini-3.1-flash-image-preview with claude-sonnet-4-5. Image prompt: Vintage 1990s novelty t-shirt screen print in deep red ink on worn mustard-yellow cotton fabric, featuring a classic View-Master stereoscopic toy viewer held by cartoon hands in center, with simple wireframe beach scene visible through eyepieces, bold retro text reading AR/VR arched across top, aged cotton texture with minor stains, humorous nostalgic charm, simple cartoon outlines, slightly imperfect printed look
Bro is speed running latent space and conjuring 3d words into existence. The branching approach is intereresting. Need to try.
https://x.com/bilawalsidhu/status/2030090631675097507
Building a new kind of map of the world becomes easier every day
https://x.com/bilawalsidhu/status/2031164408865366378
Experimenting with blending 3d gaussian splats + geospatial 3d tiles You get the best of both worlds — the insanely rich detail of your own custom 3d captures, perfectly grounded in the spatial 3d context of google’s global photogrammetry coverage Another step towards
https://x.com/bilawalsidhu/status/2031544590478590321
Just watched the most satisfying assembly timelapse ever… a fully 3D-printable Raptor 3 engine that snaps together with magnets?! @AnDuong_91 from @Morethan3D_Ltd (X) just dropped this masterpiece: by far the most detailed and accurate 1:8 scale Raptor 3 model out there.
https://x.com/IlirAliu_/status/2029633916554924061
OnlineX: Unified Online 3D Reconstruction and Understanding with Active-to-Stable State Evolution https://xiac20.github.io/OnlineX/
shader effects for 3d gaussian splats are a fun way to add reality bending effects to your 3d scans. made using sparkjs which is an open source 3dgs renderer for threejs.
https://x.com/bilawalsidhu/status/2030526450500018342
Simply adding Gaussian noise to LLMs (one step–no iterations, no learning rate, no gradients) and ensembling them can achieve performance comparable to or even better than standard GRPO/PPO on math reasoning, coding, writing, and chemistry tasks. We call this algorithm RandOpt.
https://x.com/yule_gan/status/2032482266773926281
Track4World: Future of Motion Reconstruction https://jiah-cloud.github.io/Track4World.github.io/
ETCH https://boqian-li.github.io/ETCH/
ETCH: Generalizing Body Fitting to Clothed Humans via Equivariant Tightness”” TL;DR: predicts the hidden body shape beneath clothing by learning SE(3)-equivariant cloth-to-body mappings and tightness vectors for robust clothed human fitting.
https://x.com/Almorgand/status/2029591053209096695
Track4World: Feedforward World‑centric Dense 3D Tracking of All Pixels”” TL;DR: feed‑forward model that predicts pixel‑level 2D and 3D dense flows for holistic world‑centric 3D tracking from monocular video, outperforming prior flow and tracking baselines.
https://x.com/Almorgand/status/2031060671064891647
Look at how crispy this 3d gaussian splat came out! The reflections in the painting have me floored. Captured with a Sony a7iii camera and 18mm lens. I posed the imagery using epic’s reality capture and trained it with lichtfeld studio. Lichtfeld is my new favorite way to
https://x.com/bilawalsidhu/status/2029776346742690004
OnlineX: Unified Online 3D Reconstruction and Understanding”” TL;DR: a feed‑forward online 3D reconstruction model that uses an active‑to‑stable state evolution to process streaming RGB images and build temporally consistent 3D geometry + semantic fields in real time.
https://x.com/Almorgand/status/2029620582258520124
if you know 3d tools, you’re gonna love this video 10/10 no notes – blender is now the king; houdini is the only challenger
https://x.com/bilawalsidhu/status/2029385972467601862
ID-LoRA: Identity-Driven Audio-Video Personalization https://id-lora.github.io/
Leave a Reply