AI Paper Atlas

Research map / 3D vision / Human pose estimation

Simple Baselines for Human Pose Estimation and Tracking

Bin Xiao, Haiping Wu, Yichen Wei

ECCV 2018 · 3D vision › Human pose estimation · approach: Heatmap and coordinate pose encoding · 2,290 citations (OpenAlex)

Read the paper

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 three deconvolutional layers instead of two (ablation)ResNet-50 deconv head · COCO val2017AP: 67.9 -> 70.4Improved
increase deconv kernel size from 2 to 4 (ablation)ResNet-50 deconv head · COCO val2017AP: 70.1 -> 70.4Improved
use ResNet-101 instead of ResNet-50 (ablation)simple baseline · COCO val2017AP: 70.4 -> 71.4Improved
use ResNet-152 instead of ResNet-50 (ablation)simple baseline · COCO val2017AP: 70.4 -> 72.0Improved
increase input size from 128x96 to 256x192 (ablation)simple baseline ResNet-50 · COCO val2017AP: 60.6 -> 70.4Improved
increase input size from 256x192 to 384x288 (ablation)simple baseline ResNet-50 · COCO val2017AP: 70.4 -> 72.2Improved
deconvolution head vs CPN with OHKM at same backbone and sizeCPN ResNet-50 · COCO val2017AP: 69.4 -> 70.4Improved
add joint propagation with optical flow (R-FCN detector) (ablation)tracking baseline · PoseTrack valMOTA: 57.6 -> 61.4Improved
add joint propagation with optical flow (FPN-DCN detector) (ablation)tracking baseline · PoseTrack valMOTA: 59.8 -> 62.1Improved
add joint propagation with optical flow (ResNet-152) (ablation)tracking baseline · PoseTrack valMOTA: 62.0 -> 64.8Improved
replace bbox similarity with flow-based pose similarity (single frame) (ablation)tracking with joint propagation · PoseTrack val (FPN-DCN)MOTA: 62.1 -> 62.4Improved
use multi-frame flow similarity instead of single-frame (ablation)tracking with joint propagation · PoseTrack val (FPN-DCN)MOTA: 62.4 -> 62.9Improved

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