AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

Robust Motion Segmentation From Pairwise Matches

Federica Arrigoni, Tomas Pajdla

ICCV 2019 · Detection and segmentation › Open-set and few-shot detection · approach: Amodal and open-world instance segmentation · 14 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
replace spanning-tree baseline with mode voting over redundant pairwise segmentationspairwise segmentation with RPA · Hopkins155 (all sequences)misclassification error (%): 3.79 -> 1.37Improved
replace spanning-tree baseline with mode voting over redundant pairwise segmentationspairwise segmentation with RPA · Hopkins12mean misclassification error (%): 7.45 -> 4.33Improved
replace spanning-tree baseline with mode votingpairwise segmentation with RPA · Hopkins12median misclassification error (%): 2.16 -> 0.38Improved
use Mode instead of the baseline on indoor sequencespairwise segmentation with RPA · Indoor sequencesmisclassification error (%): 0.76 vs 0.95; 3.80 vs 2.30No clear effect
use Mode instead of the baseline on indoor sequencespairwise segmentation with RPA · Indoor sequence Penguinclassified points (%): 33.95 -> 69.17Improved
use Mode instead of the baseline on helicopterpairwise segmentation with RPA · Outdoor sequence helicopterclassified points (%): 45.93 -> 80.82Improved
use Mode instead of the baseline on helicopterpairwise segmentation with RPA · Outdoor sequence helicoptermisclassification error (%): 0.78 -> 2.01Hurt
use Mode instead of track-based Subset on helicopterpairwise matches input · Outdoor sequence helicoptermisclassification error (%): 16.81 -> 2.01Improved
use Mode instead of track-based Subset on noisy indoor Penguinpairwise matches input · Indoor sequence Penguinmisclassification error (%): 32.27 -> 0.76Improved
increase fraction of mismatched correspondences up to 60%Mode · cars1 simulatedmisclassification error (%): around 0% up to 60% mismatchesNo clear effect
increase mismatches above 10% for track-based Subset and RSIMSubset and RSIM · cars1 simulatedmisclassification error (%)Hurt
compute tracks with QuichMatch instead of StableSfM (ablation)Subset and RSIM · indoor sequencesmisclassification error (%): 99.59 -> 70.11 classifiedNo clear effect

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