Hand Shape Research Articles

Saliency-Driven Hand Gesture Recognition Incorporating Histogram of Oriented Gradients (HOG) and Deep Learning.

Hand gesture recognition is a vital means of communication to convey information between humans and machines. We propose a novel model for hand gesture recognition based on computer vision methods and compare results based on images with complex scenes. While extracting skin color information is an efficient method to determine hand regions, complicated image backgrounds adversely affect recognizing the exact area of the hand shape. Some valuable features like saliency maps, histogram of oriented gradients (HOG), Canny edge detection, and skin color help us maximize the accuracy of hand shape recognition. Considering these features, we proposed an efficient hand posture detection model that improves the test accuracy results to over 99% on the NUS Hand Posture Dataset II and more than 97% on the hand gesture dataset with different challenging backgrounds. In addition, we added noise to around 60% of our datasets. Replicating our experiment, we achieved more than 98% and nearly 97% accuracy on NUS and hand gesture datasets, respectively. Experiments illustrate that the saliency method with HOG has stable performance for a wide range of images with complex backgrounds having varied hand colors and sizes.

3D hand mesh recovery through inverse kinematics from a monocular RGB image

Recovering 3D hand mesh from a monocular RGB image has a wide range of application scenarios such as VR/AR. The parametric hand model provides a good geometric prior to the shape of hand, and is commonly used to recover the 3D hand mesh. However, the rotation parameters of hand model are not easy to learn, which influences the accuracy of model-based methods. To address this problem, we take advantage of the inverse kinematic chains of hand to derive an analytical method, which can convert hand joint locations into rotation parameters. By integrating such analytical method into the neural network, we propose an end-to-end learnable model named IKHand to recover the 3D hand mesh. IKHand comprises detection module and mesh generation module. Detection module predicts the 3D hand keypoints while mesh generation module takes these keypoints to generate the 3D hand mesh. Experimental results show that our proposed method can generate impressive and robust 3D hand meshes under several challenging conditions, and can achieve competitive results on FreiHAND dataset as well as RHD.

Closed‐Loop Optimization of Soft Sensor Morphology Using 3D Printing of Electrically Conductive Hydrogel

Soft sensing technologies provide a novel alternative for state estimation in wearables and robotic systems. They allow one to capture intrinsic state parameters in a highly conformable manner. However, due to the nonlinearities in the materials that make up a soft sensor, it is difficult to develop accurate models of these systems. Consequently, design of these soft sensors is largely user defined or based on trial and error. Since these sensors conform and take the shape of the sensing body, these issues are further exacerbated when they are installed. Herein, a framework for the automated design optimization of soft sensors using closed‐loop 3D printing of a recyclable hydrogel‐based sensing material is presented. The framework allows direct printing of the sensor on the sensing body using visual feedback, evaluates the sensor performance, and iteratively improves the sensor design. Following preliminary investigations into the material and morphology parameters, this is demonstrated through the optimization of a sensorized glove which can be matched to specific tasks and individual hand shapes. The glove's sensors are tuned to respond only to particular hand poses, including distinguishing between two similar tennis racket grip techniques.

Intuitive mapping between nonsymbolic quantity and observed action across development

Adults’ concurrent processing of numerical and action information yields bidirectional interference effects consistent with a cognitive link between these two systems of representation. This link is in place early in life: infants create expectations of congruency across numerical and action-related stimuli (i.e., a small [large] hand aperture associated with a smaller [larger] numerosity). Although these studies point to a developmental continuity of this mapping, little is known about the later development and thus how experience shapes such relationships. We explored how number–action intuitions develop across early and later childhood using the same methodology as in adults. We asked 3-, 6-, and 8-year-old children, as well as adults, to relate the magnitude of an observed action (a static hand shape, open vs. closed, in Experiment 1; a dynamic hand movement, opening vs. closing, in Experiment 2) to either a small or large nonsymbolic quantity (numerosity in Experiment 1 and numerosity and/or object size in Experiment 2). From 6 years of age, children started performing in a systematic congruent way in some conditions, but only 8-year-olds (added in Experiment 2) and adults performed reliably above chance in this task. We provide initial evidence that early intuitions guiding infants’ mapping between magnitude across nonsymbolic number and observed action are used in an explicit way only from late childhood, with a mapping between action and size possibly being the most intuitive. An initial coarse mapping between number and action is likely modulated with extensive experience with grasping and related actions directed to both arrays and individual objects.

Gender Classification from Fingerprint Using Hybrid CNN-SVM

Gender classification is used in numerous applications such as biometrics, criminology, surveillance, HCI, and business profiling. Although biometric factors like gait, face, hand shape, and iris have been used to classify people into genders, the majority of research has focused on facial traits due to their more recognisable qualities. This research employs fingerprints to classify gender, with the intention of being relevant for future studies. Several methods for gender classification utilising fingerprints have been presented in the literature, including ANN, KNN, Naive Bayes, the Gaussian mixture model, and deep learning-based classifiers. Although these classifiers have shown good classification accuracy, gender classification remains an unexplored field of study that necessitates the development of new approaches to enhance recognition accuracy, computation, and running time. In this paper, a CNN-SVM hybrid framework for gender classification from fingerprints is proposed, where preprocessing, feature extraction, and classification are the three main components. The main goal of this study is to use CNN to extract fingerprint information. These features are then sent to an SVM classifier to determine gender. The hybrid model's performance measures are examined and compared to those of the conventional CNN model. Using a CNN-SVM hybrid model, the accuracy of gender classification based on fingerprints was 99.25%.

Corresponding measurement-based patternmaking method for customized gloves to support smart wearables

This study developed a corresponding measurement-based patternmaking (CMP) method for customized gloves applicable to different hand shapes to support the manufacturing of smart wearables. This method can be used directly to implement a 2D block pattern of gloves by hand measuring and 3D scanning alone. Our customized glove prototype was manufactured considering the glove’s functionality and usability and in accordance with the protocol designed, targeting individual participants. The CMP method for customized gloves was verified by checking the function and functional fit of the customized glove prototype through a wear test that compared the prototype with a commercial glove. In the process, the rules of the existing CMP method for customized leggings were adjusted for the protocol to comply with the structural specificity of the hand. Accordingly, a guideline for customized gloves’ patternmaking was proposed to help future developers. The originality of the CMP method lies in the pattern construction of a perfectly matched size, a side panel structure of C-curve fingers, and patterns for a trapezoid hand proximal part structure. Furthermore, the newly devised CMP method for customized gloves can be used as a baseline for smart wearable devices to offer high functionality and usability for individuals with unique hand shapes who need customized gloves.

The Hand of Irulegi Basque-Iberism return after seventy years of dismissal

The Hand of Irulegi is a Iberian bronze plate with a hand shape which is scripted with Iberian-Tartesian signs and has been found in an archaeological site (at Mt Irulegi) close to Pamplona, Navarre (Spain). It was probably fabricated at the 1st century BC to be hanged at the outside house door. Part os the Iberian signs have been officially transcripted and translated by official scholars by using Basque meaning of Iberian wording. It implies that Basque- Iberism returns to official scholars after about 70 years of dismissal. In this work, we have proposed a transcription and translation of the full four lines of the inscription by using phonetic-semantic similarities between Basque and Iberian wording and Iberian-Tartessian signary. Translation has come out to be in the Hospitium Iberian striking (to Romans) custum/tradition which consisted of a warm invitation and reception to foreigners to be hosted in the house. Also, it is now possible that Iberian-Tartessian rock scripts at Canary Islands may also be studied by official Iberian scholars and also the possible early origins of Iberian–Tartessian signary on the Lineal Megalith Scripts found in rocks in Mediterranean and Atlantic EuroAfrica. The main conclusion about “The Irulegi Hand” Iberian writing meaning is that it is within the frame of foreigners invitation to be hosted Iberian tradition or “Hospitium”. However, work on the exact meaning is not yet finished.

Plastic-Reconstructive Hand Surgery: a Multidisciplinary Procedure

The present study aims to provide a detailed and accurate summary of reconstructive aesthetic hand surgery in medical terms. This specialized surgical intervention comprehensively addresses the functional and aesthetic alterations affecting the hand, through complex and personalized techniques.Aesthetic reconstructive hand surgery is indicated in a variety of clinical situations, including hand trauma, congenital malformations, degenerative diseases, soft tissue defects, tumors and post-surgical reconstruction. Each case requires a thorough and personalized evaluation to determine the most appropriate surgical approach and to achieve satisfactory results.The anatomical complexity of the hand, consisting of bones, joints, tendons, muscles, nerves and blood vessels, requires in-depth knowledge on the part of the surgeon. The multidisciplinary approach, in collaboration with specialists in plastic surgery, orthopedic surgery, occupational therapy, physiotherapy and rehabilitative medicine, is crucial to the success of the intervention.Aesthetic reconstructive hand surgery seeks to restore hand function, improve aesthetic appearance and ultimately improve patients' quality of life. The focus is on repair and reconstruction of affected tissues, correction of deformities, stabilization of joints, revision of scars, and optimization of hand shape and contour

Synthetic Smartwatch IMU Data Generation from In-the-wild ASL Videos

The scarcity of training data available for IMUs in wearables poses a serious challenge for IMU-based American Sign Language (ASL) recognition. In this paper, we ask the following question: can we "translate" the large number of publicly available, in-the-wild ASL videos to their corresponding IMU data? We answer this question by presenting a video to IMU translation framework (Vi2IMU) that takes as input user videos and estimates the IMU acceleration and gyro from the perspective of user's wrist. Vi2IMU consists of two modules, a wrist orientation estimation module that accounts for wrist rotations by carefully incorporating hand joint positions, and an acceleration and gyro prediction module, that leverages the orientation for transformation while capturing the contributions of hand movements and shape to produce realistic wrist acceleration and gyro data. We evaluate Vi2IMU by translating publicly available ASL videos to their corresponding wrist IMU data and train a gesture recognition model purely using the translated data. Our results show that the model using translated data performs reasonably well compared to the same model trained using measured IMU data.

The spatio-temporal architecture of everyday manual behavior

In everyday activities, humans move alike to manipulate objects. Prior works suggest that hand movements are built by a limited set of basic building blocks consisting of a set of common postures. However, how the low dimensionality of hand movements supports the adaptability and flexibility of natural behavior is unknown. Through a sensorized glove, we collected kinematics data from thirty-six participants preparing and having breakfast in naturalistic conditions. By means of an unbiased analysis, we identified a repertoire of hand states. Then, we tracked their transitions over time. We found that manual behavior can be described in space through a complex organization of basic configurations. These, even in an unconstrained experiment, recurred across subjects. A specific temporal structure, highly consistent within the sample, seems to integrate such identified hand shapes to realize skilled movements. These findings suggest that the simplification of the motor commands unravels in the temporal dimension more than in the spatial one.

Cursor control through hand shape and Z dimension movements to assist target acquisition in large displays

Cursor control through hand shape and Z dimension movements to assist target acquisition in large displays

3D hand reconstruction with both shape and appearance from an RGB image

Modeling 3D hands with geometry details and appearance can increase perceptual immersion and realism in many applications. However, traditional 3D representations such as meshes and voxel grids can not represent high-quality hand shapes with their property of the discrete surface and fixed topology. To address these problems, we introduce implicit function for more detailed shape reconstruction, and design a Structure-aware Signed Distance Function (S-SDF) to reconstruct hand shape in arbitrary resolution. In this way, our method not only focuses on independent fingers but also keeps the relationship between fingers. Such implicit function of 3D representation associated with part semantics and structure of the hand is central to generating realistic hand shapes of diverse variations. Meanwhile, in order to avoid time-consuming and laborious 3D annotating of texture, we present a self-supervised appearance synthesis approach. Extensive experiments on widely used datasets demonstrate that the proposed method reconstructs more realistic hands compared with previous methods.

Patterns of variability of the shape of the human hand

The differences between the hands of men and women are mainly observed in the difference in the ratio of the lengths of the index and ring fingers (ratio 2d:4d) or in the difference in the proportional ratios of longitudinal and transverse dimensions, identified using classical morphometry methods, which give only indirect ideas about variations in the shape of the hand. The object of this study was digital images of radiographs of the right hands of 50 men and 50 women, on which 20 landmarks were located, the configuration and two-dimensional coordinates of which were studied using geometric morphometry methods. The predominance of the general variability of the hand’s shape was associated with a combined multidirectional transformation in the space of the elements of the II-V rays of the hand relative to the longitudinal axis with simultaneous compression or stretching of the shape relative to the transverse axis. At the same time, men have a stretching of the shape of the hand from the IV-V rays and compression from the II-III rays, while women have reverse changes. The relationship between the shape and size of the hand, regardless of gender, is minimal – 5.82% and 3.93% of hand allometry were detected in men and women, respectively. This study shows that the shape of the hand is markedly different in men and women, which indicates a significant sexual dimorphism affecting this trait. Based on the detected sexual differences, it is possible to distinguish the male and female morphological type of the hands.

Real Time Indian Sign Language Detection System

Sign language, as a different form of the communication language, is important to large groups of people in society. There are different signs in each sign language with variability in hand shape, motion profile, and position of the hand, face, and body parts contributing to each sign. A challenging area of computer vision research is the recognition of visual sign languages. In recent years, deep learning techniques have significantly improved the many models that have been suggested by various academics. We examine the deep learning-based vision-based models for sign language recognition. An important advancement in enhancing communication between the deaf and the overall populace is a real-time sign language detector. We are glad to present the development and application of a model for recognizing sign language based on a Convolutional neural network (CNN). We utilized a pre-trained SSD mobile net V2 architecture trained on our own dataset in Order to apply transfer learning to the task

Geometric Morphometric Analysis of the Human Hand Morphology Based on X-ray Imaging

Currently, in humans, the ratio of the index and ring fingers (2D:4D ratio) determines the morphological type of the hand: type 1 or radial (2D>4D), type 2 or neutral (2D=4D), type 3 or ulnar (2D<4D). However, the above differences do not allow for morphological verification of the hand shape as a whole.The aim of the study was to inspect the hand shapes depending on its morphology based on radiographical images with geometric morphometric methods.Material and methods. Geometric morphometry included an analysis of the right hand shape described by 20 landmarks located on digital radiographic images (anteriorposterior projection) in 50 males and 50 females. Each hand was classified according to whether the index finger was longer, equal to, or shorter than the ring finger by visual assessment.Results. The study revealed differences between the hand shapes in the transverse direction, namely, the hands of the ulnar type (U) were wider, while the hands of the radial type (R) were narrower. The hands of neutral shape, unlike the other two types, were more oval in shape. In average, the hand shape depended on the morphological type. The most pronounced shape transformations were due to the geometry of the metacarpal bones in the transverse direction, and due to the geometry of the II, IV and V fingers in the longitudinal direction. Intergroup differences in width were revealed between the group of hands of radial (R)-neutral (N) types and ulnar (U) type, and in length between the group of hands of radial (R)-ulnar (U) types and neutral (N) type. The allometry of the hand shape was minimal.Conclusion. The results of the study demonstrate the potentials of geometric morphometry for analysis of the human hand shapes and their variability depending on the morphological type.

Multi-cue temporal modeling for skeleton-based sign language recognition.

Sign languages are visual languages used as the primary communication medium for the Deaf community. The signs comprise manual and non-manual articulators such as hand shapes, upper body movement, and facial expressions. Sign Language Recognition (SLR) aims to learn spatial and temporal representations from the videos of the signs. Most SLR studies focus on manual features often extracted from the shape of the dominant hand or the entire frame. However, facial expressions combined with hand and body gestures may also play a significant role in discriminating the context represented in the sign videos. In this study, we propose an isolated SLR framework based on Spatial-Temporal Graph Convolutional Networks (ST-GCNs) and Multi-Cue Long Short-Term Memorys (MC-LSTMs) to exploit multi-articulatory (e.g., body, hands, and face) information for recognizing sign glosses. We train an ST-GCN model for learning representations from the upper body and hands. Meanwhile, spatial embeddings of hand shape and facial expression cues are extracted from Convolutional Neural Networks (CNNs) pre-trained on large-scale hand and facial expression datasets. Thus, the proposed framework coupling ST-GCNs with MC-LSTMs for multi-articulatory temporal modeling can provide insights into the contribution of each visual Sign Language (SL) cue to recognition performance. To evaluate the proposed framework, we conducted extensive analyzes on two Turkish SL benchmark datasets with different linguistic properties, BosphorusSign22k and AUTSL. While we obtained comparable recognition performance with the skeleton-based state-of-the-art, we observe that incorporating multiple visual SL cues improves the recognition performance, especially in certain sign classes where multi-cue information is vital. The code is available at: https://github.com/ogulcanozdemir/multicue-slr.

EFFECT OF HAND ANTHROPOMETRY INDICES ON THE MEASUREMENT OF HAND GRIP STRENGTH USING A HANDHELD DYNAMOMETER IN YOUNG ASIAN MALES

Hand grip strength is a measure of one’s muscle strength of the hand and forearm. It also indicates one’s general physical fitness. Hand grip strength is measured using a handheld dynamometer and this procedure can be affected by one’s hand anthropometry. Hand size and shape can be affected by one’s race and ethnicity and thus identifying the specific hand anthropometric measurement that would be related to hand grip strength for a particular ethnicity is important. Thus, this study investigated the relationship between different hand anthropometric measurements with hand grip strength using a handheld dynamometer in South Asian healthy young male adults. Full hand length, middle finger length, palm width and wrist thickness in 52 young male participants aged 18-25 years old were measured. Hand grip strength was measured using a handheld dynamometer. Pearson correlation coefficient was used to study the relationship of each anthropometric variable with hand grip strength. Our data showed, palm width significantly correlates with hand grip strength (r=0.461; p=0.00059). Middle finger length has a weak positive correlation (r=0.260; p=0.063), while full hand length and wrist thickness has no correlation with hand grip strength (r=0.048; p=0.734) and (r=0.094; p=0.509) respectively. Our results show that of the four anthropometry measurements of the hand used in this study, palm width has a strong association with hand grip strength, and it is a statistically significant determinant of hand grip strength in South Asian healthy male young adults. It can also be concluded that middle finger length and wrist thickness do not play a role in affecting hand grip strength.<p> </p><p><strong> Article visualizations:</strong></p><p><img src="/-counters-/soc/0028/a.php" alt="Hit counter" /></p>

Sign Language Translator Based on Raspberry Pi Camera Using The Haar Cascade Classifier Method

Sign language is the main tool of communication for people with hearing impairments. Communication is very limited and difficult to understand between normal people who do not know sign language, so an interpreter is needed. Where not everyone, even a few normal people, learns sign language, especially the Indonesian Sign Language System (SIBI). Motion Detection is an important subject in the field of computer vision, which is used by many systems. Today's Internet of Things is very helpful and facilitates daily human activities. An internet network allows a device to be controlled from a considerable distance. This study described a sign language translator tool for the deaf and speech impaired using a raspberry-pi camera and displayed it on the other device monitor. This system was built using the Python programming language and the OpenCV Library. The system is using Haar Cascade Classifier algorithm, where there will be data on all hand shapes based on the letters to be translated. This application uses the OpenCV library and Visual Studio Code IDE software connected to the Raspberry Pi Camera. The publisher will send data to other devices using the MQTT Broker to connect and display detection results to other device monitors wirelessly using a local network. The research was conducted at various distances between the hand and the webcam, from 30cm to 150cm. The research results using the Haar Cascade Classifier method to detect sign language obtained an accuracy of 82%.

Deciphering the roles of cell shape and Fat and Dachsous planar polarity in arranging the Drosophila apical microtubule network through quantitative image analysis.

In epithelial cells, planar polarization of subapical microtubule networks is thought to be important for both breaking cellular symmetry and maintaining the resulting cellular polarity. Studies in the Drosophila pupal wing and other tissues have suggested two alternative mechanisms for specifying network polarity. On one hand, mechanical strain and/or cell shape have been implicated as key determinants; on the other hand, the Fat-Dachsous planar polarity pathway has been suggested to be the primary polarizing cue. Using quantitative image analysis in the pupal wing, we reassess these models. We found that cell shape was a strong predictor of microtubule organization in the developing wing epithelium. Conversely, Fat-Dachsous polarity cues do not play any direct role in the organization of the subapical microtubule network, despite being able to weakly recruit the microtubule minus-end capping protein Patronin to cell boundaries. We conclude that any effect of Fat-Dachsous on microtubule polarity is likely to be indirect, via their known ability to regulate cell shape.

New combined DT-CWT and HOG descriptor for static and dynamic hand gesture recognition.

In recent years, researchers have been focusing on developing Human-Computer Interfaces that are fast, intuitive, and allow direct interaction with the computing environment. One of the most natural ways of communication is hand gestures. In this context, many systems were developed to recognize hand gestures using numerous vision-based techniques, these systems are highly affected by acquisition constraints, such as resolution, noise, lighting condition, hand shape, and pose. To enhance the performance under such constraints, we propose a static and dynamic hand gesture recognition system, which utilizes the Dual-Tree Complex Wavelet Transform to produce an approximation image characterized by less noise and redundancy. Subsequently, the Histogram of Oriented Gradients is applied to the resulting image to extract relevant information and produce a compact features vector. For classification, we compare the performance of three Artificial Neural Networks, namely, MLP, PNN, and RBNN. Random Decision Forest and SVM classifiers are also used to ameliorate the efficiency of our system. Experimental evaluation is performed on four datasets composed of alphabet signs and dynamic gestures. The obtained results demonstrate the efficiency of the combined features, for which the achieved recognition rates were comparable to the state-of-the-art.