AI Paper Atlas

Research map / 3D vision / Depth and stereo

Fisheye3R: Adapting Unified 3D Feed-Forward Foundation Models to Fisheye Lenses

Ruxiao Duan, Erin Hong, Dongxu Zhao, Eric Turner, Alex Wong, Yunwen Zhou

ECCV 2026 · 3D vision › Depth and stereo · approach: Cost-volume multi-view stereo · 7 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 calibration tokens trained with self-supervised learning on unlabeled perspective imagesMapAnything · KITTI360 fisheyeChamfer distance: 1.301 -> 1.015Improved
add annotated fisheye data (SL+) to trainingMapAnything · KITTI360 fisheyeChamfer distance: 1.016 -> 0.575Improved
add supervised perspective labels (SL) over SSLMapAnything · ScanNet++ fisheyeChamfer distance: 0.054 -> 0.054No clear effect
apply Fisheye3R SL+ to VGGTVGGT · ADT fisheyeChamfer distance: 0.208 -> 0.059Improved
add masked attention to calibration tokens (ablation)Fisheye3R · hybrid ScanNet++ sequencesaccuracy across perspective ratiosImproved
train tokens on fully fisheye sequences without masked attention (ablation)Fisheye3R · hybrid ScanNet++ sequencesperspective accuracyHurt
train tokens on mixed sequences without masked attention (ablation)Fisheye3R · hybrid ScanNet++ sequencesfisheye accuracyHurt
insert calibration tokens K=1 instead of 8 (ablation)Fisheye3R · ScanNet++ MapAnythingChamfer distanceNo clear effect
calibrate all 24 encoder layers instead of the last 12 (ablation)Fisheye3R · ScanNet++ MapAnythingdepth Rel: 0.181 vs 0.180No clear effect
use calibration tokens instead of LoRA r=32 (ablation)MapAnything · ScanNet++ fisheyepose/FoV vs depth/point: RPEt 0.134 vs 0.137; Rel 0.171 vs 0.170Mixed
add masked attention at inference (ablation)Fisheye3R · MapAnything speedFPS: 31.1 -> 29.1Hurt
apply tokens trained with Kannala-Brandt to other projection modelspi3 · Stanford2D3DSChamfer distance: 0.264 -> 0.095 (MEI)Improved

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