SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2110.00620 · ICCV · 2021

SPEC: Seeing People in the Wild with an Estimated Camera

Otmar Hilliges, Michael Black, Joachim Tesch, Muhammed Kocabas, Chun-Hao Huang, Lea Müller

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

Input image (b) SOTA method (c) Our results Focal length = 755 px Camera pitch = 33.4 o Camera roll = -0.6 o Figure 1: State-of-the-art methods for 3D human pose and shape estimation from images (such as HMR trained with EFT [26] data) struggle with imagery containing perspective effects. (b) In part, this is due to the use of the standard weak perspective camera. (c) SPEC learns to estimate perspective camera parameters and uses these to regress more accurate 3D poses.

For agents

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

get_harvested_code_for_paper("2110.00620")
get_code_for_paper("2110.00620")
have("2110.00620")

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

curl https://syntology.ai/api/ran/2110.00620.json

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

[![Syntology run record](https://syntology.ai/api/ran/2110.00620.svg)](https://syntology.ai/paper/2110.00620)

Connect an agent — have() is free.