Edifidio Ada Byron
Ed. Ada Byron
DIIS
DIIS
DIIS
DIIS
Ed. Ada Byron
DIIS
DIIS
DIIS
DIIS

Noticias de actualidad

Seminario: 4D Vision from Images: Data-based vs. Model-based Learning
Este viernes 10, de 17:00 a 19:00 en el seminario A.25 Ed. Ada Byron tendrá lugar el seminario que se detalla a continuación. La actividad está enmarcada dentro del programa de invitación de profesores externos.

4D Vision from Images: Data-based vs. Model-based Learning

Antonio Agudo, Investigador Postdoctoral

Instituto de Robotica e Informatica Industrial (IRI)(CSIC)

Abstract #1

This paper addresses the problem of 3D human pose estimation from a single image. We follow a standard two-step pipeline by first detecting the 2D position of the N body joints, and then using these observations to infer 3D pose. For the first step, we use a recent CNN-based detector. For the second step, most existing approaches perform 2N-to- 3N regression of the Cartesian joint coordinates. We show that more precise pose estimates can be obtained by representing both the 2D and 3D human poses using N ⇥N distance matrices, and formulating the problem as a 2D-to-3D distance matrix regression. For learning such a regressor we leverage on simple Neural Network architectures, which by construction, enforce positivity and symmetry of the predicted matrices. The approach has also the advantage to naturally handle missing observations and allowing to hypothesize the position of non-observed joints. Quantitative results on Humaneva and Human3.6M datasets demonstrate consistent performance gains over state-of-the-art. Qualitative evaluation on the images in-the-wild of the LSP dataset, using the regressor learned on Human3.6M, reveals very promising generalization results.

Abstract #2

We present an approach to reconstruct the 3D shape of multiple deforming objects from incomplete 2D trajectories acquired by a single camera. Additionally, we simultaneously provide spatial segmentation (i.e., we identify each of the objects in every frame) and temporal clustering (i.e., we split the sequence into primitive actions). This advances existing work, which only tackled the problem for one single object and non-occluded tracks. In order to handle several objects at a time from partial observations, we model point trajectories as a union of spatial and temporal subspaces, and optimize the parameters of both modalities, the non-observed point tracks and the 3D shape via augmented Lagrange multipliers. The algorithm is fully unsupervised and results in a convex formulation which does not require initialization. We thoroughly validate the method on challenging scenarios with several human subjects performing different activities which involve complex motions and close interaction. We show our approach achieves state-of-the-art 3D reconstruction results, while it also provides spatial and temporal segmentation.

Última modificación: 02/06/2015 - 16:09

Fecha: Lunes 1 de Junio, 18h
Lugar: Aula A03, Edificio Ada Byron, Campus Rio Ebro

Última modificación: 02/06/2015 - 16:09

El Máster Universitario de Ingeniería en Informática de la Universidad de Zaragoza organiza el I Ciclo de Conferencias sobre la Gestión de la Innovación TIC, coordinadas por los profesores del Departamento de Informática e Ingeniería de Sistemas: Manuel González Bedia, María Villarroya Gaudó, y Fº Javier Zarazaga Soria.

La innovación y la investigación TIC son clave para que las empresas se doten de elementos diferenciales para conseguir ventajas competitivas. Si tu ámbito profesional es éste, o esperas que lo sea en un futuro, asiste a estas conferencias (son gratis).

Última modificación: 02/06/2015 - 16:09

El próximo 29 de mayo, a las 13:00, en el Seminario 21 del edificio Ada Byron de la EINA, se presentará el Máster Universitario en Ingeniería en Informática ante alumnos y empresas.

El máster ha sido diseñado para que su realización sea compatible con el trabajo a tiempo parcial en empresa o colaboración con un grupo de investigación.