AI Paper Atlas

Research map / 3D vision / 3D generation

ROAR-3D: Routing Arbitrary Views for High-Fidelity 3D Generation

Hanxiao Sun, Mingxin Yang, Shuhui Yang, Zebin He, Xintong Han, Hongbo Fu, Chunchao Guo, Wenhan Luo

ECCV 2026 · 3D vision › 3D generation · approach: Diffusion-based novel view synthesis · 4 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add token-wise view router instead of concatenating view tokens (ablation)ROAR-3D stage 1 · Anyview-200CD: 39.412 -> 28.143Improved
add token-wise view router (ablation)ROAR-3D stage 1 · Anyview-200F1(0.05): 66.3 -> 73.5Improved
add dual-stream attention (ablation)ROAR-3D stage 1 · Anyview-200CD: 28.143 -> 26.732Improved
add dual-stream attention (ablation)ROAR-3D stage 1 · Anyview-200Uni-I: 0.335 -> 0.339Improved
add orientation perturbation (ablation)ROAR-3D stage 1 · Anyview-200CD: 26.732 -> 25.372Improved
add orientation perturbation (ablation)ROAR-3D stage 1 · Anyview-200F1(0.05): 75.1 -> 76.8Improved
add stage 2 geometric refinementROAR-3D · Held-out test setCD: 25.372 -> 21.039Improved
add stage 2 geometric refinementROAR-3D · Held-out test setF1(0.05): 76.8 -> 81.6Improved
use ROAR-3D instead of strongest baseline ReconViaGenROAR-3D stage 2 · Held-out test setCD: 32.055 -> 21.039 (34% reduction)Improved
condition on multiple views instead of single viewHunyuan3D 2.0 (MV variant) · Held-out test setCD: 35.421 -> 33.954No clear effect
increase test-time input views beyond training range (1-4 to 8)ROAR-3D · Anyview-200 subsetqualitative geometryImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy