AI Paper Atlas

Research map / 3D vision / 3D generation

KV-Tracker: Real-Time Pose Tracking with Transformers

Marwan Taher, Ignacio Alzugaray, Kirill Mazur, Xin Kong, Andrew Davison

CVPR 2026 · 3D vision › 3D generation · approach: Monocular 3D object reconstruction · 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
cache keys/values from keyframes and track single query frames instead of recomputing all-to-all attention (ablation)pi3 multi-view model · Runtime analysis (308x308)FPS: 30 FPS up to 50 frames vs quadratic slowdownImproved
use KV-cache tracking instead of streaming memory (TTT3R)KV-Tracker · TUM RGB-DATE RMSE (m): 0.132 -> 0.098Improved
use KV-cache tracking instead of TTT3RKV-Tracker · 7-ScenesATE RMSE (m): 0.143 -> 0.059Improved
KV-Tracker vs sparse DPVOKV-Tracker · TUM RGB-DATE RMSE (m): 0.095 -> 0.098No clear effect
use KV-Tracker instead of TTT3R at lower resolutionKV-Tracker · ARCTICATE RMSE (m): 0.303 -> 0.228Improved
raise input resolution from 308 to 518 (2D bbox) (ablation)KV-Tracker · OnePoserecall 3cm-3deg (%): 52.8 -> 69.3Improved
raise input resolution from 308 to 518 (2D bbox) (ablation)KV-Tracker · OnePose LowTexturerecall 5cm-5deg (%): 83.3 -> 94.4Improved
use dilated 2D bboxes instead of segmentation masks (ablation)KV-Tracker @518 · OnePose and OnePose LowTexturerecall (%): 92.1 -> 92.9 (OnePose 5cm); 10.7 -> 5.3 (OnePose 1cm)Mixed
online map vs baselines' offline 3D reconstruction (strict threshold)KV-Tracker · OnePoserecall 1cm-1deg (%): 51.1 -> 10.7Hurt
online KV-Tracker vs OnePose++ with offline maps (coarse threshold)KV-Tracker @518 2D bbox · OnePose LowTexturerecall 5cm-5deg (%): 72.1 -> 94.4Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 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