Single-Shot Object Detector Based on Attention Mechanism

2019 ◽  
Author(s):  
Xinxin Wang ◽  
Hongwei Zhu
Keyword(s):  
Attention Mechanism ◽  
Single Shot

scholarly journals Object Detection Algorithm Based on Multiheaded Attention

2019 ◽  
Vol 9 (9) ◽  
pp. 1829 ◽  
Author(s):  
Jie Jiang ◽  
Hui Xu ◽  
Shichang Zhang ◽  
Yujie Fang
Keyword(s):  
Object Detection ◽  
Linear Interpolation ◽  
Detection Algorithm ◽  
Attention Mechanism ◽  
Visual Object ◽  
Single Shot ◽  
Object Class ◽  
Feature Information ◽  
Base Network ◽  
Detector Model

This study proposes a multiheaded object detection algorithm referred to as MANet. The main purpose of the study is to integrate feature layers of different scales based on the attention mechanism and to enhance contextual connections. To achieve this, we first replaced the feed-forward base network of the single-shot detector with the ResNet–101 (inspired by the Deconvolutional Single-Shot Detector) and then applied linear interpolation and the attention mechanism. The information of the feature layers at different scales was fused to improve the accuracy of target detection. The primary contributions of this study are the propositions of (a) a fusion attention mechanism, and (b) a multiheaded attention fusion method. Our final MANet detector model effectively unifies the feature information among the feature layers at different scales, thus enabling it to detect objects with different sizes and with higher precision. We used the 512 × 512 input MANet (the backbone is ResNet–101) to obtain a mean accuracy of 82.7% based on the PASCAL visual object class 2007 test. These results demonstrated that our proposed method yielded better accuracy than those provided by the conventional Single-shot detector (SSD) and other advanced detectors.

scholarly journals SSD7-FFAM: A Real-Time Object Detection Network Friendly to Embedded Devices from Scratch

2021 ◽  
Vol 11 (3) ◽  
pp. 1096
Author(s):  
Qing Li ◽  
Yingcheng Lin ◽  
Wei He
Keyword(s):  
Object Detection ◽  
Real Time ◽  
Large Scale ◽  
Feature Fusion ◽  
Contextual Information ◽  
Attention Mechanism ◽  
Detection Accuracy ◽  
Single Shot ◽  
Feature Maps ◽  
Embedded Devices

The high requirements for computing and memory are the biggest challenges in deploying existing object detection networks to embedded devices. Living lightweight object detectors directly use lightweight neural network architectures such as MobileNet or ShuffleNet pre-trained on large-scale classification datasets, which results in poor network structure flexibility and is not suitable for some specific scenarios. In this paper, we propose a lightweight object detection network Single-Shot MultiBox Detector (SSD)7-Feature Fusion and Attention Mechanism (FFAM), which saves storage space and reduces the amount of calculation by reducing the number of convolutional layers. We offer a novel Feature Fusion and Attention Mechanism (FFAM) method to improve detection accuracy. Firstly, the FFAM method fuses high-level semantic information-rich feature maps with low-level feature maps to improve small objects’ detection accuracy. The lightweight attention mechanism cascaded by channels and spatial attention modules is employed to enhance the target’s contextual information and guide the network to focus on its easy-to-recognize features. The SSD7-FFAM achieves 83.7% mean Average Precision (mAP), 1.66 MB parameters, and 0.033 s average running time on the NWPU VHR-10 dataset. The results indicate that the proposed SSD7-FFAM is more suitable for deployment to embedded devices for real-time object detection.

An Attention Mechanism Extension of Automatic HTML Generation from Web Page Design Images

2020 ◽  
Vol 140 (12) ◽  
pp. 1393-1401
Author(s):  
Hiroki Chinen ◽  
Hidehiro Ohki ◽  
Keiji Gyohten ◽  
Toshiya Takami
Keyword(s):  
Attention Mechanism ◽  
Web Page ◽  
Web Page Design ◽  
Page Design

Functional stereotactic radiosurgery involving a dedicated linear accelerator and gamma unit: a comparison study

2004 ◽  
pp. 373-380 ◽  
Author(s):  
Timothy D. Solberg ◽  
Steven J. Goetsch ◽  
Michael T. Selch ◽  
William Melega ◽  
Goran Lacan ◽  
...  
Keyword(s):  
Stereotactic Radiosurgery ◽  
Linear Accelerator ◽  
Treatment Plan ◽  
Single Shot ◽  
Submillimeter Range ◽  
Primate Model ◽  
Content Type ◽  
Stereotactic Frame ◽  
Targeting Ability ◽  
Fine Print

Object. The purpose of this work was to investigate the targeting and dosimetric characteristics of a linear accelerator (LINAC) system dedicated for stereotactic radiosurgery compared with those of a commercial gamma knife (GK) unit. Methods. A phantom was rigidly affixed within a Leksell stereotactic frame and axial computerized tomography scans were obtained using an appropriate stereotactic localization device. Treatment plans were performed, film was inserted into a recessed area, and the phantom was positioned and treated according to each treatment plan. In the case of the LINAC system, four 140° arcs, spanning ± 60° of couch rotation, were used. In the case of the GK unit, all 201 sources were left unplugged. Radiation was delivered using 3- and 8-mm LINAC collimators and 4- and 8-mm collimators of the GK unit. Targeting ability was investigated independently on the dedicated LINAC by using a primate model. Measured 50% spot widths for multisource, single-shot radiation exceeded nominal values in all cases by 38 to 70% for the GK unit and 11 to 33% for the LINAC system. Measured offsets were indicative of submillimeter targeting precision on both devices. In primate studies, the appearance of an magnetic resonance imaging—enhancing lesion coincided with the intended target. Conclusions. Radiosurgery performed using the 3-mm collimator of the dedicated LINAC exhibited characteristics that compared favorably with those of a dedicated GK unit. Overall targeting accuracy in the submillimeter range can be achieved, and dose distributions with sharp falloff can be expected for both devices.

Sign in / Sign up

Export Citation Format

Share Document