Fwd: Seminari (dimarts 22, 16h, UOC)

[Xavier BINEFA VALLS]

dieu-me qui vol venir. Jo hi aniré

------------------------------

Xavier Binefa
Director DTIC

www.upf.edu/etic
cmtech.upf.edu

xavier...@upf.edu

Inici del missatge reenviat:

De: Àgata Lapedriza Garcia <alape...@uoc.edu>

Tema: Seminari (dimarts 22, 16h, UOC)

Data: 18 de diciembre de 2015 21:02:50 GMT+01:00

Per a: xavier...@upf.edu

Hola Xavier, com anem? Suposo que amb ganes de vacances com tothom :)

T'informo que el proper dimarts dia 22 a les 16h l'Adrià Recasens (estudiant de doctorat al MIT supervisat per l'Antonio Torralba) ens farà una xerrada a la UOC (Rambla del Poblenou 156) que crec que et pot interessar. Concretament ens parlarà del treball "Where are they looking" (un treball sobre Gaze following amb deep learning) que ha presentat recentment a NIPS. 

Estàs convidat a la xerrada i pots fer extensiva la invitació als estudiants o col·laboradors que et sembli convenient. 

T'agrairia que si decidiu venir m'ho diguéssis abans del dimarts per saber més o menys quants serem.

Et copio al final del mail l'abstract i la bio de l'Adrià.

Fins ara,

Agata Lapedriza

Estudis d'Informàtica, Multimèdia i Telecomunicació

Universitat Oberta de Catalunya

http://www.cvc.uab.es/~agata/

Abstract:

Humans have the remarkable ability to follow the gaze of other people to identify what they are looking at. Following eye gaze, or gaze-following, is an important ability that allows us to understand what other people are thinking, the actions they are performing, and even predict what they might do next. Despite the importance of this topic, this problem has only been studied in limited scenarios within the computer vision community.  

In this project, we propose a deep neural network-based approach for gaze-following and a new benchmark dataset for thorough evaluation. Given an image and the location of a head, our approach follows the gaze of the person and identifies the object being looked at. Our deep network is able to discover how to extract head pose and gaze orientation, and to select objects in the scene that are in the predicted line of sight and likely to be looked at (such as televisions, balls and food). The quantitative evaluation shows that our approach produces reliable results, even when viewing only the back of the head. While our method outperforms several baseline approaches, we are still far from reaching human performance on this task. Overall, we believe that gaze-following is a challenging and important problem that deserves more attention from the community. 

 

Short bio:

Adrià Recasens is second year PhD student in computer vision at the Computer Science and Artificial Intelligence Laboratory (CSAIL) of the Massachusetts Institute of Technology advised by Professor Antonio Torralba. His research interests range on various topics in computer vision and machine learning. He is focusing most his research on automatic gaze-following. He received a Telecommunications Engineer's Degree and a Mathematics Licentiate Degree from the Universitat Politècnica de Catalunya.

[Pares Burgues, Narcis]

A mi m'interessa molt per tot el que estem fent amb nens amb autisme.

Salut,

N

Prof. Narcis Pares Burgues
Universitat Pompeu Fabra

Sent from Android smartphone

El dia 18/12/2015 21:20, "Xavier BINEFA VALLS" <xavier...@upf.edu> va escriure:

--
Heu rebut aquest missatge perquè esteu subscrit al grup "CMTech UPF" de Google Groups.
Per cancel·lar la subscripció a aquest grup i deixar de rebre els seus missatges, envieu un correu electrònic a cmtech-upf+...@googlegroups.com.
Per veure més opcions, visiteu https://groups.google.com/d/optout.