DenseAttentionSeg: Segment hands from interacted objects using depth input

Bo, Zi-Hao; Zhang, Hao; Yong, Jun-Hai; Gao, Hao; Xu, Feng

doi:10.1016/j.asoc.2020.106297

Cited by 10 publications

(6 citation statements)

References 32 publications

(30 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…At present, the attention mechanism is widely used in the field of deep learning based on CNNs, which indicates that the attention mechanism is an effective way to improve the performance of deep CNNs. The attention mechanism can be used to readjust feature maps generated by some layers of a neural network, which makes it able to detect specific channel or spatial features 28 , 29 . In HSI classification, the attention mechanism is roughly divided into the channel attention mechanism and the spatial attention mechanism.…”

Section: Methodsmentioning

confidence: 99%

“…The attention mechanism can be used to readjust feature maps generated by some layers of a neural network, which makes it able to detect specific channel or spatial features. 28,29 In HSI classification, the attention mechanism is roughly divided into the channel attention mechanism and the spatial attention mechanism. Channel attention involves modeling the spatial context information from each channel feature and then strengthening the important channel weights to improve classification accuracy.…”

Section: Spatial-attention 3d and Channel-atttention 3dmentioning

confidence: 99%

See 1 more Smart Citation

Multi-scale hybrid three-dimensional-two-dimensional-attention boosted convolutional neural network for hyperspectral image classification

Wang

et al. 2023

J. Appl. Rem. Sens.

View full text Add to dashboard Cite

Hyperspectral image (HSI) classification is a hot topic in the field of remote sensing applications. However, due to the high-dimensional and extensive spectral and spatial information of HSIs, effective feature extraction is difficult, which leads to a lower accuracy of HSI classification. In this work, a convolutional neural network (CNN) approach based on a multiscale hybrid 3D-2D-attention mechanism (MHAC) is proposed. Linear discriminant analysis dimension reduction is performed for HSIs, and the central pixel and its adjacent pixels are input into the network as a whole. Based on a multi-scale three-dimensional convolutional neural network (3D CNN) and a two-dimensional convolutional neural network (2D CNN), fine spectral and spatial features are automatically extracted, which improves the network classification capability. At the same time, the channel attention mechanism is embedded in the 2D CNN, and the spatial attention mechanism is embedded in the 3D CNN, which can assign different weights to the retrieved features, effectively utilizing the spatial spectral information to optimize features and improve the feature extraction ability. The experimental results demonstrate that the MHAC achieves an accuracy of 98.72%, 97.60%, and 97.90% on the WHU-Hi dataset and 98.93% on the Indian-Pines dataset based on only 100 samples. The MHAC model proposed in this paper has a good feature extraction ability that effectively reduces the impact of highdimensional data and small sample problems on classification and improves the accuracy of HSIs classification.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Spatial-attention 3d and Channel-atttention 3dmentioning

confidence: 99%

Multi-scale hybrid three-dimensional-two-dimensional-attention boosted convolutional neural network for hyperspectral image classification

Wang

et al. 2023

J. Appl. Rem. Sens.

View full text Add to dashboard Cite

show abstract

“…Currently, the attention mechanism has been successfully applied to the area of computer vision based on convolutional neural networks. The attention mechanism can be used to readjust feature maps generated by some layers of a neural network, which make it able to detect specific channel or spatial features [ 45 , 46 ]. The attention mechanism can be roughly divided into spatial attention and channel attention [ 34 ].…”

Section: Methodsmentioning

confidence: 99%

Spatial–Spectral Feature Refinement for Hyperspectral Image Classification Based on Attention-Dense 3D-2D-CNN

Zhang

Feng-yuan

Feng

et al. 2020

Sensors

View full text Add to dashboard Cite

Convolutional neural networks provide an ideal solution for hyperspectral image (HSI) classification. However, the classification effect is not satisfactory when limited training samples are available. Focused on “small sample” hyperspectral classification, we proposed a novel 3D-2D-convolutional neural network (CNN) model named AD-HybridSN (Attention-Dense-HybridSN). In our proposed model, a dense block was used to reuse shallow features and aimed at better exploiting hierarchical spatial–spectral features. Subsequent depth separable convolutional layers were used to discriminate the spatial information. Further refinement of spatial–spectral features was realized by the channel attention method and spatial attention method, which were performed behind every 3D convolutional layer and every 2D convolutional layer, respectively. Experiment results indicate that our proposed model can learn more discriminative spatial–spectral features using very few training data. In Indian Pines, Salinas and the University of Pavia, AD-HybridSN obtain 97.02%, 99.59% and 98.32% overall accuracy using only 5%, 1% and 1% labeled data for training, respectively, which are far better than all the contrast models.

show abstract

“…The main types of research related to hand-tracking are connected to creating custom hand-tracking mechanisms and, for example, comparing different algorithms and their accuracy [4,5]. Additionally, some UI studies focus on designing different methods for object-grabbing using hand-tracking (mid-air grabbing) in VR [6]. Masurovsky et al [7] compare two similar camera-based hand-tracking interfaces using Leap Motion with a traditional controller solution utilising Oculus Touch controllers.…”

Section: Related Workmentioning

confidence: 99%

A Comparison of Gesture and Controller-based User Interfaces for 3D Design Reviews in Virtual Reality

Nyyssönen¹,

Helle²,

Lehtonen³

et al. 2022

Proceedings of the Annual Hawaii International Conference on System Sciences

View full text Add to dashboard Cite

We present a custom virtual reality (VR) hand-tracking user interface developed in Unity, and compare it with a similar industry level application (CTRL Elements) utilising controllers. Both applications afford spawning, relocating, reorienting, and rescaling 3D models within a virtual environment in addition to horizontal, vertical, and rotational user movement options. The interfaces are tested with Oculus Quest 2 VR headset, which is also responsible for the hand-tracking. Our main motivation is to gain an insight into whether hand-tracking UIs could provide added value to industrial design review solutions. We also pinpoint the major advantages and flaws in the controllerless approach and report which of our gesture-based controls are the most intuitive and usable.

show abstract

DenseAttentionSeg: Segment hands from interacted objects using depth input

Cited by 10 publications

References 32 publications

Multi-scale hybrid three-dimensional-two-dimensional-attention boosted convolutional neural network for hyperspectral image classification

Multi-scale hybrid three-dimensional-two-dimensional-attention boosted convolutional neural network for hyperspectral image classification

Spatial–Spectral Feature Refinement for Hyperspectral Image Classification Based on Attention-Dense 3D-2D-CNN

A Comparison of Gesture and Controller-based User Interfaces for 3D Design Reviews in Virtual Reality

Contact Info

Product

Resources

About