AI Paper Atlas

Research map / 3D vision / Depth and stereo

Pixel-Perfect Visual Geometry Estimation

Gangwei Xu, Haotong Lin, Hongcheng Luo, Haiyang Sun, Bing Wang, Guang Chen, Sida Peng, Hangjun Ye, Xin Yang

TPAMI 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
add semantics prompt (SP-DiT) to vanilla pixel-space DiT (ablation)DiT · NYUv2AbsRel: 22.5 -> 4.8Improved
add REPA token alignment to vanilla DiT (ablation)DiT · NYUv2AbsRel: 22.5 -> 17.6Improved
add Cascade DiT (coarse-to-fine patch size) to SP-DiT (ablation)SP-DiT · NYUv2AbsRel: 4.8 -> 4.3Improved
add Cascade DiT to SP-DiT (ablation)SP-DiT · RTX 4090inference time: 0.20 -> 0.14Improved
add Cascade DiT to SP-DiT (ablation)SP-DiT · ScanNetAbsRel: 6.2 -> 4.5Improved
use DINOv2 encoder instead of MAE for semantics (ablation)SP-DiT · NYUv2AbsRel: 6.4 -> 4.8Improved
use MoGe 2 encoder instead of Depth Anything v2 for semantics (ablation)SP-DiT · NYUv2AbsRel: 4.3 -> 3.3Improved
add reference-guided token propagation (RGTP) (ablation)SP-DiT (DAv2) · NYUv2 videoAbsRel: 12.2 -> 7.6Improved
replace SP-DiT with multi-view SC-DiT (VGGT) with RGTP (ablation)PPVD · NYUv2 videoAbsRel: 7.6 -> 4.5Improved
use pi^3 instead of VGGT as multi-view semantic encoder (ablation)SC-DiT + RGTP · NYUv2 videoAbsRel: 4.5 -> 3.8Improved
use PPVD instead of Video Depth Anythingvideo depth · ScanNetAbsRel: 8.9 -> 3.7Improved
use pixel-space diffusion instead of VAE latent space (GT(VAE) reconstruction) (ablation)depth representation · Hypersim edge-awareChamfer DistanceImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 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