AI Paper Atlas

Research map / 3D vision / Depth and stereo

From None to All: Self-Supervised 3D Reconstruction via Novel View Synthesis

Ranran Huang, Weixun Luo, Ye Mao, Krystian Mikolajczyk

CVPR 2026 · 3D vision › Depth and stereo · approach: Foundation models for monocular geometry · 2 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
use NAS3R instead of SelfSplat or SPFSplat* for self-supervised 2-view NVSNAS3R · RE10K 2-view NVSPSNR: 21.306 -> 23.130 vs SPFSplat*Improved
use NAS3R instead of SPFSplat* for pose estimationNAS3R · RE10K 2-view poseoverall AUC@10: 23.9 -> 51.0Improved
use depth-based local-to-global Gaussian formulation without pretrained priorsNAS3R · BlendedMVS depthrel: 0.255 -> 0.206 vs SPFSplat*Improved
train on RE10K+DL3DV instead of RE10K only (ablation)NAS3R · ScanNet++ pose AUC@10AUC@10: 19.2 -> 24.5Improved
train on DL3DV instead of RE10K only (ablation)NAS3R · ScanNet++ pose AUC@10AUC@10: 19.2 -> 19.1No clear effect
increase number of context views from 2 to 10 (ablation)NAS3R · RE10K NVSPSNR: 23.130 -> 27.093Improved
initialize with MASt3R or VGGT pretrained priors (ablation)NAS3R · RE10K 2-view NVSPSNR: 23.130 -> 25.888 (VGGT)Improved
provide GT intrinsics to the backbone (NAS3R-I) (ablation)NAS3R (VGGT) · RE10K 2-view poseAUC@5: 62.4 -> 67.7Improved
initialize supervised fine-tuning from NAS3R self-supervised weights (ablation)BlendedMVS supervised training · BlendedMVS depthrel: 0.177 -> 0.119Improved
self-supervised NAS3R versus supervised-from-scratch training on BlendedMVS pose (ablation)NAS3R · BlendedMVS poseAUC@5: 10.7 -> 29.8Improved
NAS3R(VGGT) versus NAS3R(MASt3R) with priors (ablation)NAS3R · RE10K 2-view poseAUC@5: 63.8 (MASt3R) vs 62.4 (VGGT)No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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