Human pose completion in partial body camera shots

Ruben Tous, Jordi Nin, Laura Igual

Producció científica: Article en revista indexadaArticleAvaluat per experts

Resum

Many actual images contain partial body camera shots, in which a significant part of the body is not visible. This issue is especially prevalent in film images, where less than 10% are full-body shots. Most 2D human pose estimation methods return incomplete poses when applied to partial body images. This lack of completeness becomes a problem in some situations, for example, when the 2D pose is converted to a 3D pose by a two-stage 3D human pose estimation method since most of these methods require a complete pose to work. This article proposes a new technique, called CompletePose, consisting of completing the missing keypoints when 2D human pose estimation methods are applied to images with partial body camera shots. A Conditional Generative Adversarial Network is used to obtain a complete and plausible pose, realistic enough to predict a 3D pose with a two-stage 3D human pose estimation method. A complete empirical validation has been carried out with the Human3.6 M dataset and a new dataset, called CHARADE, specially built and made public for reproducibility and benchmarking for this research. Quantitative evaluation employing the Fréchet distance shows that the approach manages to approximate the actual data distribution. The qualitative evaluation shows that the completed poses enable obtaining plausible 3D poses from images previously intractable.

Idioma originalAnglès
Nombre de pàgines11
RevistaJournal of Experimental and Theoretical Artificial Intelligence
Data online anticipadade jul. 2023
DOIs
Estat de la publicacióPublicació electrònica prèvia a la impressió - de jul. 2023

Fingerprint

Navegar pels temes de recerca de 'Human pose completion in partial body camera shots'. Junts formen un fingerprint únic.

Com citar-ho