Saved in:
Bibliographic Details
Main Authors: Subramanian, Sanjay, Ng, Evonne, Müller, Lea, Klein, Dan, Ginosar, Shiry, Darrell, Trevor
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2405.03689
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916737593638912
author Subramanian, Sanjay
Ng, Evonne
Müller, Lea
Klein, Dan
Ginosar, Shiry
Darrell, Trevor
author_facet Subramanian, Sanjay
Ng, Evonne
Müller, Lea
Klein, Dan
Ginosar, Shiry
Darrell, Trevor
contents Language is often used to describe physical interaction, yet most 3D human pose estimation methods overlook this rich source of information. We bridge this gap by leveraging large multimodal models (LMMs) as priors for reconstructing contact poses, offering a scalable alternative to traditional methods that rely on human annotations or motion capture data. Our approach extracts contact-relevant descriptors from an LMM and translates them into tractable losses to constrain 3D human pose optimization. Despite its simplicity, our method produces compelling reconstructions for both two-person interactions and self-contact scenarios, accurately capturing the semantics of physical and social interactions. Our results demonstrate that LMMs can serve as powerful tools for contact prediction and pose estimation, offering an alternative to costly manual human annotations or motion capture data. Our code is publicly available at https://prosepose.github.io.
format Preprint
id arxiv_https___arxiv_org_abs_2405_03689
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Pose Priors from Language Models
Subramanian, Sanjay
Ng, Evonne
Müller, Lea
Klein, Dan
Ginosar, Shiry
Darrell, Trevor
Computer Vision and Pattern Recognition
Computation and Language
Language is often used to describe physical interaction, yet most 3D human pose estimation methods overlook this rich source of information. We bridge this gap by leveraging large multimodal models (LMMs) as priors for reconstructing contact poses, offering a scalable alternative to traditional methods that rely on human annotations or motion capture data. Our approach extracts contact-relevant descriptors from an LMM and translates them into tractable losses to constrain 3D human pose optimization. Despite its simplicity, our method produces compelling reconstructions for both two-person interactions and self-contact scenarios, accurately capturing the semantics of physical and social interactions. Our results demonstrate that LMMs can serve as powerful tools for contact prediction and pose estimation, offering an alternative to costly manual human annotations or motion capture data. Our code is publicly available at https://prosepose.github.io.
title Pose Priors from Language Models
topic Computer Vision and Pattern Recognition
Computation and Language
url https://arxiv.org/abs/2405.03689