Skip to Main content Skip to Navigation
Preprints, Working Papers, ...

Thin-Slicing for Pose: Learning to Understand Pose without Explicit Pose Estimation

Suha Kwak 1, 2 Minsu Cho 2, 1 Ivan Laptev 1
1 WILLOW - Models of visual object recognition and scene understanding
DI-ENS - Département d'informatique de l'École normale supérieure, Inria Paris-Rocquencourt, CNRS - Centre National de la Recherche Scientifique : UMR8548
Abstract : We address the problem of learning a pose-aware, compact embedding that projects images with similar human poses to be placed close-by in the embedding space. The embedding function is built on a deep convolutional network , and trained with a triplet-based rank constraint on real image data. This architecture allows us to learn a robust representation that captures differences in human poses by effectively factoring out variations in clothing, background, and imaging conditions in the wild. For a variety of pose-related tasks, the proposed pose embedding provides a cost-efficient and natural alternative to explicit pose estimation, circumventing challenges of localizing body joints. We demonstrate the efficacy of the embedding on pose-based image retrieval and action recognition problems.
Document type :
Preprints, Working Papers, ...
Complete list of metadata
Contributor : Suha Kwak Connect in order to contact the contributor
Submitted on : Monday, December 14, 2015 - 1:54:06 AM
Last modification on : Friday, October 15, 2021 - 1:40:08 PM
Long-term archiving on: : Tuesday, March 15, 2016 - 11:11:11 AM


Files produced by the author(s)


  • HAL Id : hal-01242724, version 1


Suha Kwak, Minsu Cho, Ivan Laptev. Thin-Slicing for Pose: Learning to Understand Pose without Explicit Pose Estimation. 2015. ⟨hal-01242724v1⟩



Record views


Files downloads