ReCal3R: Reliability-Calibrated Learning Rates for Streaming 3D Reconstruction
Streaming 3D reconstruction relies on a compact recurrent scene state to process long image streams in linear time and bounded memory. However, repeated updates can gradually corrupt this state, causing reliable historical information to be overwritten by noisy or ambiguous observations. We introduce ReCal3R, a reliability-calibrated learning rate method for recurrent 3D reconstruction. Instead of directly applying a candidate learning rate, our method estimates state token reliability from the maintained scene state and uses it to calibrate a candidate learning rate derived from token alignme
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 45%francelico/PERSIST →
- LinkedLinked via arxiv author · 85%Xinze Li →
“ReCal3R: Reliability-Calibrated Learning Rates for Streaming 3D Reconstruction”
- LinkedLinked via arxiv author · 85%Yiyuan Wang →
“ReCal3R: Reliability-Calibrated Learning Rates for Streaming 3D Reconstruction”
- LinkedLinked via arxiv author · 85%Pengxu Chen →
“ReCal3R: Reliability-Calibrated Learning Rates for Streaming 3D Reconstruction”
- LinkedLinked via arxiv author · 85%Wentao Fan →
“ReCal3R: Reliability-Calibrated Learning Rates for Streaming 3D Reconstruction”
- LinkedLinked via arxiv author · 85%Weifeng Su →
“ReCal3R: Reliability-Calibrated Learning Rates for Streaming 3D Reconstruction”
- LinkedLinked via arxiv author · 85%Weisi Lin →
“ReCal3R: Reliability-Calibrated Learning Rates for Streaming 3D Reconstruction”
- LinkedLinked via arxiv author · 85%Wentao Cheng →
“ReCal3R: Reliability-Calibrated Learning Rates for Streaming 3D Reconstruction”