SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2104.08527 · ICCV · 2021

PARE: Part Attention Regressor for 3D Human Body Estimation

Otmar Hilliges, Michael Black, Muhammed Kocabas, Chun-Hao Huang

arXiv · PDF · Open in the Atlas

Code that ran

We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Image (b) SPIN [1] (c) SPIN [2] 1 2 (d) SPIN Occlusion Sensitivity Heatmap (e) PARE [1] (f) PARE [2] 1 2 (g) PARE Occlusion Sensitivity Heatmap 50 100 150 200 250 Mean 3D joint error (mm) Figure 1: Occlusion Sensitivity Analysis. Given an input image (a), a small occluding patch (shown in gray) causes SPIN [29] to fail (b,c), whereas our method (PARE) (e,f) is robust to the occluder. Sub-figures on the right show the sensitivity of SPIN (d) and PARE (g) to an occluding patch (the size of the white squares) centered at every point in the image. Warmer colors mean higher average joint error.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2104.08527")
get_code_for_paper("2104.08527")
have("2104.08527")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2104.08527.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2104.08527.svg)](https://syntology.ai/paper/2104.08527)

Connect an agent — have() is free.