Research model that builds complete 3D assets from sparse, unposed images by jointly doing reconstruction and generation.
Updated EU status checked Last shown
Sources: the vendor's own pages (linked above) and creator videos at the linked timestamps. EU status is information, not legal advice; see how we check.
UniForge3D (also referred to as Uniform 3D in some coverage) is a research model for generating complete 3D assets from a small number of unposed multi-view images of an object. It combines reconstruction of the geometry visible in the inputs with generation of unseen regions using learned shape priors.
The method addresses a coordinate mismatch between camera-relative reconstruction and canonical-space generation by working in a shared anchor-guided space that follows the anchor view's azimuth while keeping the object upright. The reconstruction and generation branches exchange features bidirectionally to predict per-view point maps and a complete sparse voxel structure. The predicted structure then conditions point-map prediction to establish pixel-to-voxel correspondences, so the most informative input view is selected for each voxel and its features guide detailed shape generation.
The authors report state-of-the-art appearance and geometry results on the Toys4K and GSO benchmarks. A YouTube review states that code and models were released, but no repository or model links were present in the provided material, so this is not confirmed here. The vendor or institution is not stated on the project page; the paper is cited as arXiv preprint 2026 by Chen et al.
Watch the moment: the exact second top creators show it. Videos load only when you press play, via youtube-nocookie.com.
Turns a few multi-angle images of an object into a full 3D model, combining reconstruction of visible parts with generation of missing geometry. Preserves more detail than compared methods; code and models released.
Closest neighbours by meaning, not just by category.
Model
Image-to-3D generator from Tsinghua and Tencent ARC Lab that builds meshes pixel-aligned to the input view.
Model · Meta
Meta research framework that aligns a frozen SAM 3D model to input images or video at inference time for 3D reconstruction.
Model
Feed-forward 3D scene reconstruction model that outputs triangle primitives directly usable as meshes for simulation.
Research
Post-training method that refines feed-forward 3D reconstruction models (Pi3, DA3, VGGT) on unlabeled video.
Model
Research method that reconstructs a 4D Gaussian Splatting human from a single casual monocular video.
Model
Feed-forward system that turns an RGB image or video into editable, textured per-object 3D meshes in under a minute.
