In the recent years, the demand for unsupervised annotation tools to annotate and classify large audiovisual datasets has grown considerably. One of these tasks is concretely addressed on TV broadcast videos, to determine who and when appears in a video sequence. This work is aimed on exploring deep learning methods for face feature extraction and the implementation of a verification system in order to boost the performance of person recognition tasks. A comparison between different identification methods that have been developed during this project is made, with the aim of evaluating their performance and to conclude which ones are the best. Finally, a comparison between the results obtained with this system and the one proposed by the 201...
The annotation of video streams by automatic content analysis is a growing field of research. The po...
Abstract. The Repere challenge is a project aiming at the evaluation of systems for supervised and u...
This paper describes a system to identify people in broadcast TV shows in a purely unsupervised mann...
In the recent years, the demand for unsupervised annotation tools to annotate and classify large aud...
The purpose of this project is to focus in exploring deep learning methods for multimodal person rec...
In unsupervised identity recognition in video sequences systems, which is a very active field of res...
In the recent years, the demand for video tools to automatically annotate and classify large audiovi...
The enormous amount of visual data generated nowadays creates a strong need for annotation tools to ...
Nowadays, Artificial Intelligence in computation vision is widely used, and there are constant impro...
Recogniton of peple by speech and face in TV shows. Participation in the international competition M...
Poster session: WS21 - Workshop on Information Fusion in Computer Vision for Concept RecognitionInte...
L'analyse automatique de contenu des vidéos en vue de leur annotation est un domaine de recherche en...
This paper describes a multi-modal person recognition sys-tem for video broadcast developed for part...
Treballs Finals de Grau d'Enginyeria Informàtica, Facultat de Matemàtiques, Universitat de Barcelona...
Deep learning has become part of many state-of-the-art systems in multiple disciplines (specially in...
The annotation of video streams by automatic content analysis is a growing field of research. The po...
Abstract. The Repere challenge is a project aiming at the evaluation of systems for supervised and u...
This paper describes a system to identify people in broadcast TV shows in a purely unsupervised mann...
In the recent years, the demand for unsupervised annotation tools to annotate and classify large aud...
The purpose of this project is to focus in exploring deep learning methods for multimodal person rec...
In unsupervised identity recognition in video sequences systems, which is a very active field of res...
In the recent years, the demand for video tools to automatically annotate and classify large audiovi...
The enormous amount of visual data generated nowadays creates a strong need for annotation tools to ...
Nowadays, Artificial Intelligence in computation vision is widely used, and there are constant impro...
Recogniton of peple by speech and face in TV shows. Participation in the international competition M...
Poster session: WS21 - Workshop on Information Fusion in Computer Vision for Concept RecognitionInte...
L'analyse automatique de contenu des vidéos en vue de leur annotation est un domaine de recherche en...
This paper describes a multi-modal person recognition sys-tem for video broadcast developed for part...
Treballs Finals de Grau d'Enginyeria Informàtica, Facultat de Matemàtiques, Universitat de Barcelona...
Deep learning has become part of many state-of-the-art systems in multiple disciplines (specially in...
The annotation of video streams by automatic content analysis is a growing field of research. The po...
Abstract. The Repere challenge is a project aiming at the evaluation of systems for supervised and u...
This paper describes a system to identify people in broadcast TV shows in a purely unsupervised mann...