Object Detection Based on VGG with ResNet Network

Haque, Foysal; Lim, Hye-Youn; Kang, Dae-Seong

doi:10.23919/elinfocom.2019.8706476

Cited by 57 publications

(23 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Due to its unique architecture, ResNet has enjoyed success in image classification and object detection tasks [52][53][54]. Object detection methods usually utilize the last convolution layer (conv5_x) for generating region proposals.…”

Section: Senet: Scale Estimation Networkmentioning

confidence: 99%

Multi-Scale Person Localization With Multi-Stage Deep Sequential Framework

Khan¹,

Basalamah²

2021

IJCIS

View full text Add to dashboard Cite

Person detection in real videos and images is a classical research problem in computer vision. Person detection is a nontrivial problem that offers many challenges due to several nuisances that commonly observed in natural videos. Among these, scale is the main challenging problem in various object detection tasks. To solve the scale problem, we propose a framework that estimates the scales of person's heads, as we argue that head is the only visible part in complex scenes. we propose a head detection framework that explicitly handles head scales. The framework consists of two sequential networks: (1) scale estimation network (SENet) and (2) head detection network. SENet predicts the distribution of scales from the input image in the form of histogram. Then the scale histogram adjust anchor scale set of region proposal network that generates object proposals. These objects proposals are then classified into two classes, that is, head and background by the detection network. We evaluate proposed framework on three challenging benchmark datasets. Experiment results show that proposed framework achieves state-of-the-art performance.

show abstract

Section: Senet: Scale Estimation Networkmentioning

confidence: 99%

Multi-Scale Person Localization With Multi-Stage Deep Sequential Framework

Khan¹,

Basalamah²

2021

IJCIS

View full text Add to dashboard Cite

show abstract

“…In [Alyazyah Alsuwaidi, 2017], sensors attached to clothes are used to identify the angle of the spine. however, excessive calibration for most postures is a necessity.…”

Section: Related Workmentioning

confidence: 99%

“…hence, it is essential that people in the workplace be cognizant of their sitting posture in order to reduce health risks over a long period of time. According to [Alyazyah Alsuwaidi, 2017], 76% admitted that their posture was not good. More than 90% of these people were found to be experiencing back pain on a regular basis.…”

Section: Introductionmentioning

confidence: 99%

Slouch Detection Using Open Pose Architecture

Shah¹

2020

Biosci. Biotech. Res. Comm

View full text Add to dashboard Cite

Slouch detection refers to computer vision techniques that detect human figures in images in order to determine the current posture and check if a person is sitting ergonomically. This is an important subject because following poor posture habits is known to be one of the primary causes of back pain because of its adverse influence on the transverse abdominis muscle. Maintaining good posture could help improve one's health in the long run. This paper implements slouch detection using OpenPose architecture, which first finds the relative position of joints with respect to the body. This information then helps determine whether the human subject in a given image is sitting ergonomically or not. In this paper's implementation, high accuracy was achieved and slouch detection was made more feasible by eliminating the need for several sensors.

show abstract

“…The building components of VGG are exactly the same as LeNet and AlexNet except that it is an even deeper network with more convolutional, pooling, and dense layers. VGG-16 consists of 16 weight layers: 13 convolution layers followed by pooling layers, and 3 fully connected layers [15]. It showed significant improvement to Alexnet-50 by replacing the large kernel filters 11x11 and 5x5 in the initial stage with conventional 3x3 filters one after the other [16].…”

Section: Vgg -16mentioning

confidence: 99%

Development of State of the Art AI Vision Algorithm on Xilinx Alveo U-200 FPGA Cloud and CPU+GPU Platform

Sunder¹

2020

IJERT

View full text Add to dashboard Cite

Since the start of modern computer algorithm practices, a lot of new techniques on computer vision has been developed. This results in utilizing machine learning algorithms and has provided attributes to the evolution of convolutional neural networks for building state-of-the-art object detection, segmentation and classification algorithms. These CNN can achieve human-like results in computer vision application, however with the expense of more computation. To meet machine learning application requirements on hardware deployments, various AI-Accelerated FPGA development kits have been developed along with specialized toolkits aimed at efficient optimization and deployment of the models. In theory, the FPGA solutions can have similar accuracy, better inference time and power consumption compared to the GPUs, however, it comes at the cost of limited CNN model support and additional FPGA hardware design complexity. In this thesis, an existing object detection algorithm has been studied and realtime simulation of the object detection algorithm which works under a darknet framework utilizing both CPU+GPU efficiently using CUDA by Nvidia. Implementation of GoogleNet and ResNet50 object detection algorithm on a cloudbased FPGA platform using Xilinx Vitis-AI Toolkit has been carried out. The tools utilize different strategies like model quantization and hardware architecture set up to achieve an accuracy similar to a GPU with at least 10% difference. A broad case study on hardware and software configurations made on Xilinx ALVEO U-200 FPGA for efficient deployment via the cloud has been carried out. Results of both the simulation platforms have been compared and discussed for further optimization and developments.

show abstract

Object Detection Based on VGG with ResNet Network

Cited by 57 publications

References 6 publications

Multi-Scale Person Localization With Multi-Stage Deep Sequential Framework

Multi-Scale Person Localization With Multi-Stage Deep Sequential Framework

Slouch Detection Using Open Pose Architecture

Development of State of the Art AI Vision Algorithm on Xilinx Alveo U-200 FPGA Cloud and CPU+GPU Platform

Contact Info

Product

Resources

About