DeepAirSig: End-to-End Deep Learning Based in-Air Signature Verification

Malik, Jameel; Elhayek, Ahmed; Guha, Suparna; Ahmed, Sheraz; Gillani, Amna; Stricker, Didier

doi:10.1109/access.2020.3033848

Cited by 27 publications

(20 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The authors in [13] used LMC to obtain the 3D positions of fingertips, the center of the palm and the orientation of the hand. References [14] developed an air-writing recognition scheme using 3D trajectories of fingertips acquired by an Intel RealSense 3D depth camera.…”

Section: Related Workmentioning

confidence: 99%

“…2D camera-based systems often utilize color markers on fingers to increase tracking performance since finger tracking without markers is challenging. 3D camerabased systems address the hand/finger tracking problem well simply using the depth information provided by 3D image sensors such as Kinect [12], Leap Motion Controller (LMC) [13], or Intel RealSense camera [14].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Air-Writing Recognition Based on Deep Convolutional Neural Networks

Hsieh

Chen

et al. 2021

IEEE Access

View full text Add to dashboard Cite

Air-writing recognition has received wide attention due to its potential application in intelligent systems. To date, some of the fundamental problems in isolated writing have not been addressed effectively. This paper presents a simple yet effective air-writing recognition approach based on deep convolutional neural networks (CNNs). A robust and efficient hand tracking algorithm is proposed to extract air-writing trajectories collected by a single web camera. The algorithm addresses the push-to-write problem and avoids restrictions on the users' writing without using a delimiter and an imaginary box. A novel preprocessing scheme is also presented to convert the writing trajectory into appropriate forms of data, making the CNNs trained with these forms of data simpler and more effective. Experimental results indicate that the proposed approach not only obtains much higher recognition accuracy but also reduces the network complexity significantly compared to the popular image-based methods.INDEX TERMS Air writing recognition, isolated writing, convolutional neural network (CNN), hand tracking.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Air-Writing Recognition Based on Deep Convolutional Neural Networks

Hsieh

Chen

et al. 2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…J. Malik et al [26] presented a novel method for end-toend deep learning based in-air signature verification using a depth sensor cameras. In this regard, they proposed a new medium-scale in-air signature dataset which is created using an accurate convolutional neural network (CNN) based 3D hand pose estimation algorithm.…”

Section: ) Computer Vision-based Techniquementioning

confidence: 99%

On-Air Hand-Drawn Doodles for IoT Devices Authentication During COVID-19

Elshenaway

Guirguis

2021

IEEE Access

View full text Add to dashboard Cite

In this paper, a new natural human interaction authentication method has been proposed for the Internet of Things (IoT) devices. In this method the user draws doodles on-air for authentication. On-air drawing, refers to virtually drawing free hand-drawn doodles passwords through hand gestures on the air without touching anything which is recommended during COVID-19. This work uses Google Quick Draw doodles dataset for password doodles. The proposed method is based on a usual video camera, two lightweight Convolutional Neural Networks (CNN) and Kalman filter. The first CNN for hand gestures classification to overcome dynamic hand gestures challenges on the air. The second CNN for authentication verification. Kalman filter is used to correct and smooth the drawn line path on the air. To accept the new authentication method, it must achieve two main goals security and usability. The evaluation of the usability was based on ISO 9241-11:2018 standards usability model. The results revealed that the accuracy of the proposed authentication method is 95% and, the efficiency is 94% and user satisfaction is accepted. The evaluation of the security was based on two threats related to IoT devices which are guessing and physical observation. The results showed that the password strength of the proposed authentication method is stronger than the traditional 4-digits PIN password. The proposed authentication method is also resistant to physical observation threats.

show abstract

“…T RACKING and reconstruction of human hand pose in 3D is an extensively studied computer vision problem which often arises in user authentication, augmented and virtual reality, gaming, movie production as well as human performance capture and analysis, among other fields [1], [2], [3], [4]. Accurate 3D hand tracking can facilitate gesture recognition and enable new interfaces for human-computer interaction.…”

Section: Introductionmentioning

confidence: 99%

“…HandVoxNet++ is based on 3D and graph convolutions which regresses two different representations of hand shape, namely voxelized hand shape and hand surface (Secs. [3][4][5]. The voxelized hand shape is estimated from a new voxel-to-voxel network relying on Truncated Signed Distance Field (TSDF), and establishes a one-to-one mapping between the voxelized depth map and the voxelized hand shape.…”

Section: Introductionmentioning

confidence: 99%

HandVoxNet++: 3D Hand Shape and Pose Estimation using Voxel-Based Neural Networks

Malik¹,

Shimada²,

Elhayek³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

3D hand shape and pose estimation from a single depth map is a new and challenging computer vision problem with many applications. Existing methods addressing it directly regress hand meshes via 2D convolutional neural networks, which leads to artifacts due to perspective distortions in the images. To address the limitations of the existing methods, we develop HandVoxNet++, i.e., a voxel-based deep network with 3D and graph convolutions trained in a fully supervised manner. The input to our network is a 3D voxelized-depth-map-based on the truncated signed distance function (TSDF). HandVoxNet++ relies on two hand shape representations. The first one is the 3D voxelized grid of hand shape, which does not preserve the mesh topology and which is the most accurate representation. The second representation is the hand surface that preserves the mesh topology. We combine the advantages of both representations by aligning the hand surface to the voxelized hand shape either with a new neural Graph-Convolutions-based Mesh Registration (GCN-MeshReg) or classical segment-wise Non-Rigid Gravitational Approach (NRGA++) which does not rely on training data. In extensive evaluations on three public benchmarks, i.e., SynHand5M, depth-based HANDS19 challenge and HO-3D, the proposed HandVoxNet++ achieves the state-of-the-art performance. In this journal extension of our previous approach presented at CVPR 2020, we gain 41.09% and 13.7% higher shape alignment accuracy on SynHand5M and HANDS19 datasets, respectively. Our method is ranked first on the HANDS19 challenge dataset (Task 1: Depth-Based 3D Hand Pose Estimation) at the moment of the submission of our results to the portal in August 2020.

show abstract

DeepAirSig: End-to-End Deep Learning Based in-Air Signature Verification

Cited by 27 publications

References 42 publications

Air-Writing Recognition Based on Deep Convolutional Neural Networks

Air-Writing Recognition Based on Deep Convolutional Neural Networks

On-Air Hand-Drawn Doodles for IoT Devices Authentication During COVID-19

HandVoxNet++: 3D Hand Shape and Pose Estimation using Voxel-Based Neural Networks

Contact Info

Product

Resources

About