Active learning for deep detection neural networks
Aghdam, Hamed H. (Centre de Visió per Computador (Bellaterra, Catalunya))
Gonzalez Garcia, Abel (Centre de Visió per Computador (Bellaterra, Catalunya))
Weijer, Joost van de (Centre de Visió per Computador (Bellaterra, Catalunya))
López Peña, Antonio M. (Universitat Autònoma de Barcelona. Departament de Ciències de la Computació)

Publicación:	Institute of Electrical and Electronics Engineers (IEEE), cop.2019
Descripción:	9 pàg.
Resumen:	The cost of drawing object bounding boxes (i. e. labeling) for millions of images is prohibitively high. For instance, labeling pedestrians in a regular urban image could take 35 seconds on average. Active learning aims to reduce the cost of labeling by selecting only those images that are informative to improve the detection network accuracy. In this paper, we propose a method to perform active learning of object detectors based on convolutional neural networks. We propose a new image-level scoring process to rank unlabeled images for their automatic selection, which clearly outperforms classical scores. The proposed method can be applied to videos and sets of still images. In the former case, temporal selection rules can complement our scoring process. As a relevant use case, we extensively study the performance of our method on the task of pedestrian detection. Overall, the experiments show that the proposed method performs better than random selection.
Ayudas:	Agencia Estatal de Investigación TIN2017-88709-R Agencia Estatal de Investigación TIN2016-79717-R
Nota:	Altres ajuts: as CVC members, the authors also thank the Generalitat de Catalunya CERCA Program and its ACCIO agency. Antonio thanks the financial support by ICREA under the ICREA Academia Program
Derechos:	Tots els drets reservats.
Lengua:	Anglès
Documento:	Capítol de llibre ; recerca ; Versió acceptada per publicar
Materia:	Labeling ; Task analysis ; Detectors ; Videos ; Neural networks ; Training ; Object detection
Publicado en:	2019 IEEE/CVF International Conference on Computer Vision (ICCV), 2019, p. 3671-3679, ISBN 978-1-7281-4803-8

DOI: 10.1109/ICCV.2019.00377

Postprint 10 p, 1.5 MB

El registro aparece en las colecciones:
Libros y colecciones > Capítulos de libros