Scenes as Objects, Not Primitives: Instance-Structured 3D Tokenization from Unposed Views
About
A 3D scene is understood through its objects, not the primitives that compose them. Yet feed-forward reconstruction methods output dense, unstructured sets of points or Gaussians, leaving object-level structure to be recovered after the fact. We propose a feed-forward framework that decomposes a scene into instance-structured 3D token groups directly from unposed multi-view images -- compact object-centric units from which reconstruction, segmentation, and manipulation all follow. Each token group pairs an instance token capturing entity-level identity with anchor tokens that encode local geometry and appearance, which are decoded into a set of 3D Gaussians. This two-level factorization decouples object identity from local appearance, making object instances a native interface of the representation rather than a derived product. The token groups are learned through differentiable rendering with joint reconstruction and segmentation supervision, requiring no 3D annotations. Our feed-forward model surpasses per-scene optimization baselines in class-agnostic instance segmentation while remaining competitive in novel view synthesis. Beyond these metrics, the same token groups directly unlock instance-level scene editing -- removing, translating, or inserting objects by operating on their groups -- as well as efficient open-vocabulary 3D instance retrieval, where retrieval complexity scales with the number of instances rather than primitives.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| View Reconstruction | RE10K 2-view (test) | PSNR22.85 | 13 | |
| Novel View Reconstruction | ScanNet 8 | PSNR22.41 | 6 | |
| Source view feature lifting | ScanNet 2-view | mIoU66.1 | 5 | |
| Target view feature lifting | ScanNet 2-view | mIoU65.7 | 5 | |
| Class-agnostic novel-view instance segmentation | ScanNet 8 context views | AP23.5 | 4 | |
| Novel View Reconstruction | ScanNet 2-view | PSNR25.28 | 4 |