Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition

He, Kai; Zhang, Xiangyu; Ren, Shouxin; Sun, Jian

doi:10.1007/978-3-319-10578-9_23

Cited by 1,941 publications

(1,245 citation statements)

References 41 publications

Supporting

Mentioning

1,154

Contrasting

Order By: Relevance

“…2D-CNN has been demonstrated with great promise in the field of computer vision and image processing, with applications such as image classification [38][39][40], object detection [41,42], and depth estimation from a single image [43]. The most significant advantage of 2D-CNN is that it offers a principled way to extract features directly from the raw input imagery.…”

Section: D Convolution Operationmentioning

confidence: 99%

Spectral–Spatial Classification of Hyperspectral Imagery with 3D Convolutional Neural Network

2017

View full text Add to dashboard Cite

Abstract:Recent research has shown that using spectral-spatial information can considerably improve the performance of hyperspectral image (HSI) classification. HSI data is typically presented in the format of 3D cubes. Thus, 3D spatial filtering naturally offers a simple and effective method for simultaneously extracting the spectral-spatial features within such images. In this paper, a 3D convolutional neural network (3D-CNN) framework is proposed for accurate HSI classification. The proposed method views the HSI cube data altogether without relying on any preprocessing or post-processing, extracting the deep spectral-spatial-combined features effectively. In addition, it requires fewer parameters than other deep learning-based methods. Thus, the model is lighter, less likely to over-fit, and easier to train. For comparison and validation, we test the proposed method along with three other deep learning-based HSI classification methods-namely, stacked autoencoder (SAE), deep brief network (DBN), and 2D-CNN-based methods-on three real-world HSI datasets captured by different sensors. Experimental results demonstrate that our 3D-CNN-based method outperforms these state-of-the-art methods and sets a new record.

show abstract

Section: D Convolution Operationmentioning

confidence: 99%

Spectral–Spatial Classification of Hyperspectral Imagery with 3D Convolutional Neural Network

2017

View full text Add to dashboard Cite

show abstract

“…Therefore, it was soon enhanced to learn from volumetric data represented by occupancy grids [31], [46]. Another issue tackled by the CNN community was developing solutions for obtaining invariance to different kinds of transformations [28] and scale changes [18]. In [16], an adaptive, multi-scale CNN architecture was proposed to jointly perform depth prediction, surface normal estimation and semantic labeling.…”

Section: Related Workmentioning

confidence: 99%

3D Object Detection and Recognition for Robotic Grasping Based on RGB-D Images and Global Features

Czajewski

Kołomyjec

2017

Foundations of Computing and Decision Sciences

View full text Add to dashboard Cite

Abstract. This paper describes the results of experiments on detection and recognition of 3D objects in RGB-D images provided by the Microsoft Kinect sensor. While the studies focus on single image use, sequences of frames are also considered and evaluated. Observed objects are categorized based on both geometrical and visual cues, but the emphasis is laid on the performance of the point cloud matching method. To this end, a rarely used approach consisting of independent VFH and CRH descriptors matching, followed by ICP and HV algorithms from the Point Cloud Library is applied. Successfully recognized objects are then subjected to a classical 2D analysis based on color histogram comparison exclusively with objects in the same geometrical category. The proposed two-stage approach allows to distinguish objects of similar geometry and different visual appearance, like soda cans of various brands. By separating geometry and color identification phases, the applied system is still able to categorize objects based on their geometry, even if there is no color match. The recognized objects are then localized in the three-dimensional space and autonomously grasped by a manipulator. To evaluate this approach, a special validation set was created, and additionally a selected scene from the Washington RGB-D Object Dataset was used.

show abstract

“…International Conference on Mechanical, Electronic, Control and Automation Engineering (MECAE 2017) SPPnet [4], Fast R-CNN [5], Faster R-CNN [6], etc. Among them, Faster R-CNN has the best performance, both in detection time and accuracy.…”

Section: Advances In Engineering Research (Aer) Volume 61mentioning

confidence: 99%

A New Fault Detection Algorithm for EMUs based on Deep Learning

Liu¹,

Sun²

2017

Proceedings of the 2017 International Conference on Mechanical, Electronic, Control and Automation Engineering (MECAE 2017)

View full text Add to dashboard Cite

Abstract. Health monitoring is an important task for high-speed railway EMUs. Traditional methods for detection have the problem of low detection rate or high false alarm. In this paper, a new fault detection algorithm is proposed based on deep learning. A region proposal network is applied and a network is trained based on EMU images. Initial experiments show that the proposed network can achieve 95.67% accuracy in detection, which the speed of 0.1-0.2 second per image.

show abstract

Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition

Cited by 1,941 publications

References 41 publications

Spectral–Spatial Classification of Hyperspectral Imagery with 3D Convolutional Neural Network

Spectral–Spatial Classification of Hyperspectral Imagery with 3D Convolutional Neural Network

3D Object Detection and Recognition for Robotic Grasping Based on RGB-D Images and Global Features

A New Fault Detection Algorithm for EMUs based on Deep Learning

Contact Info

Product

Resources

About