Efficient inception V2 based deep convolutional neural network for real‐time hand action recognition

Bose, Smarajit; Kumar, V. Sathiesh

doi:10.1049/iet-ipr.2019.0985

Cited by 38 publications

(26 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ablation experiments ( Supplementary Table 2) showed that the most important factors for model performance were the usage of dice loss and the addition of squeeze blocks. We experimented with other architectural blocks, namely inception [14] and residuals [15], that have been employed in similar computer vision tasks [16] but were not able to observe any significant improvements. We found that having a constant number of filters (64) across every layer of the network performs better than the typical increase/decrease in filters for every downsampling/upsampling step, respectively.…”

mentioning

confidence: 99%

deepBlink: Threshold-independent detection and localization of diffraction-limited spots

Eichenberger

Zhan

Rempfler

et al. 2020

Preprint

View full text Add to dashboard Cite

Detection of diffraction-limited spots is traditionally performed with mathematical operators designed for idealized spots. This process requires manual tuning of parameters that is time-consuming and not always reliable. We have developed deepBlink, a neural network- based method to detect and localize spots automatically and demonstrate that deepBlink outperforms state-of-the-art methods across six publicly available datasets. deepBlink is open-sourced on PyPI and GitHub (https://github.com/BBQuercus/deepBlink) as a ready-to- use command-line interface.

show abstract

mentioning

confidence: 99%

deepBlink: Threshold-independent detection and localization of diffraction-limited spots

Eichenberger

Zhan

Rempfler

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…1. YOLO-V2 overcomes the challenges faced by the other recognition systems like Single Shot Detector (SSD) [7], and the Faster Region based Convolutional Neural Network (Faster-RCNN) [6]. YOLO-V2 is the modified version of conventional YOLO architecture.…”

Section: Methodsmentioning

confidence: 99%

“…YOLO-V2 model utilizes DarkNet-19 CNN architecture as a backbone for extracting the feature vectors from the image. The YOLO-V2 CNN model has been trained and evaluated with the assistance of two benchmark datasets including the NUS Hand posture-II (NUSHP-II) dataset [19], Senz 3D hand dataset (SENZ-3D) [20] and the custom designed dataset (MITI-HD) [7]. Figure 2 illustrates the flow diagram that describes the collection of data samples and the technique used for pre-processing the hand gesture data samples.…”

Section: Methodsmentioning

confidence: 99%

“…In this paper, a YOLO-V2 based CNN architecture is proposed to increase the performance and reduce the computational time of the real-time hand action recognition system. The proposed model YOLO-V2 CNN is trained and evaluated on three datasets, namely NUSHP-II [19], SENZ-3D [20] and MITI-HD [7]. Table 1 shows the pseudo code of the system.…”

Section: Literature Reviewmentioning

confidence: 99%

“…The location of the object in visual information is represented and Classified by means of CNN algorithms known as object detectors. Hand sign recognition can be achieved by using two different methodologies namely, two stage detection system (Faster R-CNN) [6] and single stage detection systems (SSD) [7].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Precise Recognition of Vision Based Multi-hand Signs Using Deep Single Stage Convolutional Neural Network

Bose

Kumar

2021

Communications in Computer and Information Science

View full text Add to dashboard Cite

The precise recognition of multi-hand signs in real-time under dynamic backgrounds, illumination conditions is a time consuming process. In this paper, a time efficient single stage convolutional neural network (CNN) You Only Look Once (YOLO-V2) model is proposed for real-time multi-hand sign recognition. The model utilizes DarkNet-19 CNN architecture as a feature extractor. The model is trained and tested on three distinct datasets (NUSHP-II, SENZ-3D and MITI-HD). The range of IoU from 0.5 to 0.95, the model is validated using test dataset. For the MITI-HD, the YOLO-V2 CNN model obtained an average precision value of 99.10% for AP 0.5 ; 93.00% for AP 0.75 and 78.30% for AP 0.5:0.95 . The Adam Optimizer on YOLO-V2 CNN model supersedes the other optimization methods. The prediction time of YOLO-V2 CNN is obtained as 20 ms, much lower than other single-stage hand sign recognition systems.

show abstract

Convolutional neural network and its pretrained models for image classification and object detection: A survey

Jena

Nayak

Saxena

2021

Concurrency and Computation

View full text Add to dashboard Cite

At present, in the age of computers and automation of services, deep learning (DL) technology, mainly the subset of machine learning (ML) and artificial intelligence (AI), is expressively used in innumerable domains of computer vision such as data analysis, image recognition, classification, natural language processing, and many more. It has become the foremost choice of researchers as of its effectiveness in producing decent results. This paper presents detailed and analytical literature starting from the very elementary level to the recent trends of this trending technology while focusing on the most used DL model, that is, convolutional neural network and its pretrained models for image classification and object detection. It also reviews diverse existing current literature based on this. Further, a brief introduction of AI, ML, and DL has also been presented, making the foundation for the readers. As pretrained models continuously give an upper edge to DL over ML and other technologies, 23 most popular pretrained models with their architectural diagrams have also been presented. This paper aims to summarize and analyze all the concepts used to formulate DL and its models. Also, we have emphasized more on the GoogleNet models and the entire Inception modules in detail. Finally, the fascinating applications and discussion on integral components of DL have been presented. This paper will definitely draw the attention of the students and researchers working in the area of DL and its models.

show abstract

Efficient inception V2 based deep convolutional neural network for real‐time hand action recognition

Cited by 38 publications

References 25 publications

deepBlink: Threshold-independent detection and localization of diffraction-limited spots

deepBlink: Threshold-independent detection and localization of diffraction-limited spots

Precise Recognition of Vision Based Multi-hand Signs Using Deep Single Stage Convolutional Neural Network

Convolutional neural network and its pretrained models for image classification and object detection: A survey

Contact Info

Product

Resources

About