AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

E-RayZer: Self-supervised 3D Reconstruction as Spatial Visual Pre-training

Qitao Zhao, Hao Tan, Qianqian Wang, Sai Bi, Kai Zhang, Kalyan Sunkavalli, Shubham Tulsiani, Hanwen Jiang

CVPR 2026 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Invariance-based self-supervised learning · 14 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 explicit 3D Gaussians instead of RayZer's implicit latent sceneE-RayZer vs RayZer · DL3DV pose estimation, DL3DV trainingRPA@5: 0.0 -> 72.0Improved
use explicit 3D Gaussians instead of implicit latent sceneE-RayZer vs RayZer · WildRGB-D novel-view synthesis, DL3DV trainingPSNR: 25.9 -> 24.3Hurt
add visual-overlap curriculum (geometric) versus no curriculum (ablation)E-RayZer · DL3DV training, DL3DV poseRPA@5: 4.0 -> 72.0Improved
visual-overlap curriculum instead of frame-interval curriculum (ablation)E-RayZer · DL3DV training, DL3DV poseRPA@5: 56.1 -> 72.0Improved
semantic overlap instead of geometric overlap for curriculum (ablation)E-RayZer · DL3DV and 7-dataset trainingRPA@5: 73.2 vs 72.0 (DL3DV); 58.7 vs 59.9 (7-dataset)No clear effect
initialize VGGT* with E-RayZer weightsVGGT* · DL3DV in-domain poseRPA@5: 79.6 -> 87.3Improved
train on 7-dataset mix instead of DL3DV only (ablation)E-RayZer · CO3Dv2 poseRPA@5: 19.1 -> 30.3Improved
train on 7-dataset mix instead of DL3DV only (ablation)E-RayZer · DL3DV pose (in-domain)RPA@5: 72.0 -> 59.9Hurt
train on DL3DV instead of RE10K (ablation)E-RayZer · DL3DV poseRPA@5: 21.2 -> 72.0Improved
use E-RayZer encoder instead of DINOv2/DINOv3/CroCo v2/VideoMAE V2 (frozen)multi-view depth and pose probing · ScanNet++AbsRel: 0.193 (DINOv2) -> 0.116Improved
use E-RayZer encoder instead of RayZer for flowpairwise flow probing · StaticThings3DEPE: 1.105 -> 1.254Hurt
E-RayZer versus supervised VGGT* when trained on large dataE-RayZer · 7-dataset training, DL3DV poseRPA@5: 66.1 (VGGT*) vs 59.9 (E-RayZer)Hurt

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