PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
March 22, 2026Digital humanities quarterlyOpen Access

Assemblies of Points: Strategies to Art-historical Human Pose Estimation and Retrieval

View Full Paper
Ask AI
Bookmark
Share

Authors

SSStefanie Schneider

Discussion

Loading...

Member takes

Overview

This research reveals a refined method for human posture classification in art images, suggesting new approaches in visual arts.

Key Points

  • To construct a virtual space for embedding historical human figures and propose a view-invariant approach to human pose retrieval.
  • Developed a view-invariant approach to Human Pose Retrieval (HPR) with 110 art-historical reference postures.
  • Validated method effectiveness through broad-scale filtering combined with detailed individual posture analysis.
  • Utilized aggregate-level analysis of metadata and individual-level analysis of topic-specific query postures.
  • Identified key metadata-induced hotspots for human postures in art-historical images.
  • Confirmed the validity of Deep Learning-based methods through analysis of canonical forms depicted in crucifixion art.

Cite This Study

Stefanie Schneider (2026) studied this question.

synapsesocial.com/papers/69bf390ac7b3c90b18b432b3https://doi.org/10.63744/3nfsurptvbh6
View Full Paper
Ask AI
Bookmark
Share