Deep learning for visual understanding: A review

Guo, Yanming; Liu, Yu; Oerlemans, Ard; Lao, Songyang; Wu, Song; Lew, Michael S.

doi:10.1016/j.neucom.2015.09.116

Cited by 2,021 publications

(1,081 citation statements)

References 123 publications

Supporting

Mentioning

1,053

Contrasting

Unclassified

Order By: Relevance

“…The deep reinforcement learning model extends the cognitive ability of mobile robots for more complicated indoor environments in an efficient onlinelearning process continuously. As a result, this work extends a previous systematic review [7] by taking into account a different educational level. Moreover, the current systematic review expands the previous one by paying attention to the learning theories support for practical activities with robots.…”

Section: Resultssupporting

confidence: 54%

See 1 more Smart Citation

Mobile Robot Applications Grounded in Deep Learning Theories: A Review

Zheng¹

2017

IRATJ

View full text Add to dashboard Cite

Mobile robots facing a diversity of open environments and performing a variety of tasks and interactions need explicit deliberation in order to fulfill their missions. Deliberation is meant to endow a mobile robot system with extended, more adaptable and robust functionalities, as well as reduce its deployment cost. The ambition of this survey is to present a global overview of deliberation functions in mobile robots and to discuss the state of the art in deep learning theories. And the present article focuses on the study of the intelligent navigation techniques, which are capable of navigating a mobile robot autonomously in static as well as dynamic environments.

show abstract

Section: Resultssupporting

confidence: 54%

“…The feature learning algorithms, especially convolutional neural networks, have been widely used in many areas, i.e., computer vision and speech recognition [6,7]. However, it is seldom used in mobile visual navigation.…”

Section: Discussionmentioning

confidence: 99%

Mobile Robot Applications Grounded in Deep Learning Theories: A Review

Zheng¹

2017

IRATJ

View full text Add to dashboard Cite

show abstract

“…In recent years, deep learning methods have attracted significant attention [11] and have achieved revolutionary successes in various applications [12,24]. Two important structures for deep learning are CNN and RNN.…”

Section: Usage Of Cnn-rnn Frameworkmentioning

confidence: 99%

CNN-RNN: a large-scale hierarchical image classification framework

Guo¹,

Liu²,

Bakker³

et al. 2017

Multimed Tools Appl

Self Cite

123

View full text Add to dashboard Cite

Objects are often organized in a semantic hierarchy of categories, where finelevel categories are grouped into coarse-level categories according to their semantic relations. While previous works usually only classify objects into the leaf categories, we argue that generating hierarchical labels can actually describe how the leaf categories evolved from higher level coarse-grained categories, thus can provide a better understanding of the objects. In this paper, we propose to utilize the CNN-RNN framework to address the hierarchical image classification task. CNN allows us to obtain discriminative features for the input images, and RNN enables us to jointly optimize the classification of coarse and fine labels. This framework can not only generate hierarchical labels for images, but also improve the traditional leaf-level classification performance due to incorporating the hierarchical information. Moreover, this framework can be built on top of any CNN architecture which is primarily designed for leaf-level classification. Accordingly, we build a high performance network based on the CNN-RNN paradigm which outperforms the original CNN (wider-ResNet) and also the current state-of-the-art. In addition, we investigate how to utilize the CNN-RNN framework to improve the fine category classification when a fraction of the training data is only annotated with coarse labels. Experimental results demonstrate that CNN-RNN can use the coarse-labeled training data to improve the classification of fine categories, and in some cases it even surpasses the performance achieved by fully annotated training data. This reveals that, CNN-RNN can alleviate the challenge of specialized and expensive annotation of fine labels.

show abstract

“…Los métodos desarrollados para la clasificación se pueden agrupar en dos categorías: los modelos basados en la generación manual de características [24,25,32], y los modelos de aprendizaje automático de características usando técnicas de aprendizaje profundo (DL, Deep Learning) [8,11,[33][34][35][36][37][38].…”

Section: Clasificación De Peatones Sobre Imágenes En El Infrarrojo Leunclassified

“…Por otra parte, debido al reciente éxito que han presentado las técnicas de aprendizaje profundo (Deep Learning) [10,11], el principal objetivo de este trabajo es poner en marcha un método para la detección de peatones en la noche usando información visual en el infrarrojo lejano y las redes neuronales convolucionales, específicamente las arquitecturas del tipo Faster R-CNN [9,[11][12][13][14][15] para obtener un sistema competitivo que genere resultados de vanguardia comparables a los existentes en los trabajos previos. Por lo tanto, se presenta una nueva arquitectura Faster R-CNN a múlti-ples escalas, la cual es evaluada bajo los conjuntos de prueba de las bases datos CVC-09 [16] y LSIFIR [17].…”

Section: Introductionunclassified

Detección de peatones en la noche usando Faster R-CNN e imágenes infrarrojas

Bravo¹,

Flores²

2018

ings

View full text Add to dashboard Cite

En este artículo se presenta un sistema de detección de peatones en la noche, para aplicaciones en seguridad vehicular. Para este desarrollo se ha analizado el desempeño del algoritmo Faster R-CNN con imágenes en el infrarrojo lejano. Por lo que se constató que presenta inconvenientes a la hora de detectar peatones a larga distancia. En consecuencia, se presenta una nueva arquitectura Faster R-CNN dedicada a la detección en múltiples escalas, mediante dos generadores de regiones de interés (ROI) dedicados a peatones a corta y larga distancia, denominados RPNCD y RPNLD, respectivamente. Esta arquitectura ha sido comparada con los modelos para Faster R-CNN que han presentado los mejores resultados, como son VGG-16 y Resnet 101. Los resultados experimentales se han desarrollado sobre las bases de datos CVC-09 y LSIFIR, los cuales demostraron mejoras, especialmente en la detección de peatones a larga distancia, presentando una tasa de error versus FPPI de 16 % y sobre la curva Precisión vs. Recall un AP de 89,85 % para la clase peatón y un mAP de 90 % sobre el conjunto de pruebas de las bases de datos LSIFIR y CVC-09.

show abstract

Deep learning for visual understanding: A review

Cited by 2,021 publications

References 123 publications

Mobile Robot Applications Grounded in Deep Learning Theories: A Review

Mobile Robot Applications Grounded in Deep Learning Theories: A Review

CNN-RNN: a large-scale hierarchical image classification framework

Detección de peatones en la noche usando Faster R-CNN e imágenes infrarrojas

Contact Info

Product

Resources

About