AI Paper Atlas

Research map / 3D vision / Human pose estimation

iMED: A Multi-Endoscope Dataset for Surgical 3D Perception

Sierra Bonilla, Fengyi Jiang, Chinedu Nwoye, Jingpei Lu, Kailey Reardon, Humphrey Chow, Francisco Vasconcelos, Sophia Bano, Adam Schmidt, Omid Mohareri

ECCV 2026 · 3D vision › Human pose estimation · approach: Learned camera pose estimation · 0 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 depth supervision (Depth Anything) (ablation)NeRF · iMED rigid NVS, held-out endoscopePSNR: 17.24 -> 16.29Hurt
add depth supervision (FoundationStereo) (ablation)NeRF · iMED rigid NVS, held-out endoscopePSNR: 17.24 -> 15.93Hurt
add depth supervision (FoundationStereo) (ablation)3DGS · iMED rigid NVS, held-out endoscopePSNR / SSIM: PSNR 14.82 -> 15.33; SSIM 0.59 -> 0.63Improved
add depth supervision (Depth Anything) (ablation)3DGS · iMED rigid NVS, held-out endoscopePSNR / SSIM: PSNR 14.82 -> 14.77; SSIM 0.59 -> 0.63Mixed
explicit dense 3D regularization (DynOmo) vs photometric-only D-3DGSdeformable NVS · iMED deformable NVS, held-out endoscopePSNR: 14.50 -> 25.29Improved
swap depth prior from Depth Anything to FoundationStereo (ablation)DynOmo · iMED online deformable NVSPSNR: 24.16 -> 25.29Improved
swap depth prior from RAFT to FoundationStereo (ablation)Hayoz · iMED online deformable NVSPSNR: 15.66 -> 15.92No clear effect
dense foundation matcher vs sparse matcher for rotation accuracyRoMa vs GIM+LightGlue · iMED pose estimationrotation error (deg): 1.87 -> 1.79Improved
sparse learned matcher vs dense matcher for temporal consistencyGIM+LightGlue vs RoMa · iMED pose estimationtemporal drift (deg): 1.41 -> 0.41Improved
general-purpose metric depth model vs surgical-domain modelUniDepth-V2 vs Endo3R · iMED monocular depth (median scaled)Abs Rel: 0.187 -> 0.081Improved
use predicted metric scale without median scaling (ablation)UniDepth-V2 · iMED monocular depthAbs Rel: 0.081 -> 11.573Hurt

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