Patch-based 3D Human Pose Refinement

Abstract

State-of-the-art 3D human pose estimation approaches typically estimate pose from the entire RGB image in a single forward run. In this paper, we develop a post-processing step to refine 3D human pose estimation from body part patches. Using local patches as input has two advantages. First, the fine details around body parts are zoomed in to high resolution for preciser 3D pose prediction. Second, it enables the part appearance to be shared between poses to benefit rare poses. In order to acquire informative representation of patches, we explore different input modalities and validate the superiority of fusing predicted segmentation with RGB. We show that our method consistently boosts the accuracy of state-of-the-art 3D human pose methods.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 16, 2019
Accession Number
AD1154474

Entities

People

  • Alan L. Yuille
  • Qingfu Wan
  • Weichao Qiu

Organizations

  • Fudan University
  • Johns Hopkins University

Tags

DTIC Thesaurus Topics

  • Accuracy
  • Artificial Intelligence Computing
  • Artificial Intelligence Software
  • Computer Vision
  • Computing System Architectures
  • Deep Learning
  • High Resolution
  • Image Processing
  • Image Recognition
  • Image Segmentation
  • Information Science
  • Learning
  • Low Resolution
  • Motion Capture
  • Neural Networks
  • Orientation (Direction)
  • Recognition
  • Residuals
  • Training

Fields of Study

  • Computer science

Readers

  • Computational Modeling and Simulation
  • Computer Vision.