AI Paper Atlas

Research map / 3D vision / Human pose estimation

Learning in an Uncertain World: Representing Ambiguity Through Multiple Hypotheses

Christian Rupprecht, Iro Laina, Robert DiPietro, Maximilian Baust, Federico Tombari, Nassir Navab, Gregory D. Hager

ICCV 2017 · 3D vision › Human pose estimation · approach: Heatmap and coordinate pose encoding · 198 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
increase hypotheses from 1 (SHP) to 10 (10-MHP) (ablation)pose regression CNN · LSP pose estimation (oracle)strict PCP: 59.7 -> 62.8Improved
replace single prediction with 10 hypotheses (ablation)future frame CNN · NTU RGB-D put on a hat/capsharpness: 319.5 -> 419.7Improved
replace single prediction with 10 hypotheses (ablation)future frame CNN · NTU RGB-D put on a hat/capmin MSE: 960.6 -> 728.5Improved
replace single prediction with 9 hypotheses (ablation)ResNet-101 · VOC07 testmAP: 83.8 -> 85.1Improved
replace single prediction with 9 hypotheses (ablation)ResNet-101 · VOC12 valmAP: 86.9 -> 87.6Improved
replace single prediction with 5 hypotheses (ablation)ResNet-101 · COCO valmAP: 65.2 -> 67.8Improved
replace single prediction with 5 hypotheses (ablation)ResNet-101 · COCO valmAP@10: 81.0 -> 83.3Improved
increase hypotheses from 9 to 13 (ablation)ResNet-101 MHP · VOC12 valmAP: 87.6 -> 87.0Hurt
increase hypotheses from 5 to 13 (ablation)ResNet-101 MHP · VOC07 testmAP: 84.7 -> 84.7; 9-MHP peaks at 85.1Mixed
use shared-architecture 4-MHP FCN8s instead of MCL ensembleFCN8s · VOC2012 segmentationmean IoU: 69.1 -> 70.3Improved
use ResNet-101 instead of ResNet-50 or VGG-16 backbone (ablation)MHP classifier · VOC multi-labelmAP: 2-3% worse for ResNet-50 and VGG-16Improved

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