AI Paper Atlas

Research map / 3D vision / Human pose estimation

StarMap for Category-Agnostic Keypoint and Viewpoint Estimation

Xingyi Zhou, Arjun Karpur, Linjie Luo, Qixing Huang

ECCV 2018 · 3D vision › Human pose estimation · approach: Category-level object pose estimation · 75 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
use ground-truth StarMap instead of predicted (ablation)StarMap viewpoint pipeline · Pascal3D+ viewpointmean Acc@pi/6: 0.8225 -> 0.8211No clear effect
use ground-truth CanViewFeature instead of predicted (ablation)StarMap viewpoint pipeline · Pascal3D+ viewpointmean Acc@pi/6: 0.8225 -> 0.8651Improved
use ground-truth DepthMap instead of predicted (ablation)StarMap viewpoint pipeline · Pascal3D+ viewpointmean Median Error: 10.43 -> 4.7Improved
use ground-truth DepthMap instead of predicted (ablation)StarMap viewpoint pipeline · Pascal3D+ viewpointmean Acc@pi/6: 0.8225 -> 0.8637Improved
use DepthMap lifting rather than direct PnP alignmentStarMap viewpoint estimator · Pascal3D+ viewpointmean Acc@pi/6: 0.7416 -> 0.8225Improved
use DepthMap lifting rather than direct PnP alignmentStarMap viewpoint estimator · Pascal3D+ viewpointmean Acc@pi/18: 0.4643 -> 0.4818Improved
use keypoint alignment rather than category-general bin classificationviewpoint estimator · Pascal3D+ viewpointmean Acc@pi/18: 0.3502 -> 0.4818Improved
use category-specific classifiers rather than a general oneRes18 viewpoint classifier · Pascal3D+ viewpointmean Acc@pi/6: 0.7875 -> 0.8121No clear effect
use CanViewFeature instead of SIFT or Conv5 featureskeypoint classification · Pascal3D+ keypoint classificationmean accuracy: 57 -> 81Improved
use single-channel StarMap instead of 16-channel Hourglass outputHourglassNetwork · MPII (oracle ID)PCKh@0.5 total: 93.22 -> 88.17Hurt
use single-channel StarMap instead of 16-channel Hourglass outputHourglassNetwork · MPII (learned ID)PCKh@0.5 total: 88.39 -> 79.85Hurt
replace depth sub-network with DepthMap channelsZhou et al. 3D pose · Human3.6MMPJPE (mm): 64.9 -> 65.77No 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