Catalogue Search | MBRL
Are you sure you want to remove the book from the shelf?
{{itemTitle}}
333
result(s) for
"Finger spelling."
Sort by:
The handmade alphabet
Presents the handshape for each letter of the American manual alphabet accompanied by an object whose name begins with that letter.
Book
American Sign Language Alphabet Recognition by Extracting Feature from Hand Pose Estimation
Sign language is designed to assist the deaf and hard of hearing community to convey messages and connect with society. Sign language recognition has been an important domain of research for a long time. Previously, sensor-based approaches have obtained higher accuracy than vision-based approaches. Due to the cost-effectiveness of vision-based approaches, researchers have been conducted here also despite the accuracy drop. The purpose of this research is to recognize American sign characters using hand images obtained from a web camera. In this work, the media-pipe hands algorithm was used for estimating hand joints from RGB images of hands obtained from a web camera and two types of features were generated from the estimated coordinates of the joints obtained for classification: one is the distances between the joint points and the other one is the angles between vectors and 3D axes. The classifiers utilized to classify the characters were support vector machine (SVM) and light gradient boosting machine (GBM). Three character datasets were used for recognition: the ASL Alphabet dataset, the Massey dataset, and the finger spelling A dataset. The results obtained were 99.39% for the Massey dataset, 87.60% for the ASL Alphabet dataset, and 98.45% for Finger Spelling A dataset. The proposed design for automatic American sign language recognition is cost-effective, computationally inexpensive, does not require any special sensors or devices, and has outperformed previous studies.
Journal Article
Language and Reading Progress of Young Deaf and Hard-of-Hearing Children
Abstract
We examined the language and reading progress of 336 young DHH children in kindergarten, first and second grades. Trained assessors tested children’s language, reading, and spoken and fingerspelled phonological awareness in the fall and spring of the school year. Children were divided into groups based on their auditory access and classroom communication: a spoken-only group (n = 101), a sign-only group (n = 131), and a bimodal group (n = 104). Overall, children showed delays in language and reading compared to norms established for hearing children. For language, vocabulary standard scores were higher than for English syntax. Although delayed in language, children made expected gains based on hearing norms from kindergarten to second grade. Reading scores declined from kindergarten to second grade. Spoken-only and bimodal children had similar word reading and reading comprehension abilities and higher scores than sign-only children. Spoken-only children had better spoken phonological awareness and nonword reading skills than the other two groups. The sign-only and bimodal groups made similar and significant gains in ASL syntax and fingerspelling phonological awareness.
Journal Article
Deep multimodal-based finger spelling recognition for Thai sign language: a new benchmark and model composition
Video-based sign language recognition is vital for improving communication for the deaf and hard of hearing. Creating and maintaining quality of Thai sign language video datasets is challenging due to a lack of resources. Tackling this issue, we rigorously investigate a design and development of deep learning-based system for Thai Finger Spelling recognition, assessing various models with a new dataset of 90 standard letters performed by 43 diverse signers. We investigate seven deep learning models with three distinct modalities for our analysis: video-only methods (including RGB-sequencing-based CNN-LSTM and VGG-LSTM), human body joint coordinate sequences (processed by LSTM, BiLSTM, GRU, and Transformer models), and skeleton analysis (using TGCN with graph-structured skeleton representation). A thorough assessment of these models is conducted across seven circumstances, encompassing single-hand postures, single-hand motions with one, two, and three strokes, as well as two-hand postures with both static and dynamic point-on-hand interactions. The research highlights that the TGCN model is the optimal lightweight model in all scenarios. In single-hand pose cases, a combination of the Transformer and TGCN models of two modalities delivers outstanding performance, excelling in four particular conditions: single-hand poses, single-hand poses requiring one, two, and three strokes. In contrast, two-hand poses with static or dynamic point-on-hand interactions present substantial challenges, as the data from joint coordinates is inadequate due to hand obstructions, stemming from insufficient coordinate sequence data and the lack of a detailed skeletal graph structure. The study recommends integrating RGB-sequencing with visual modality to enhance the accuracy of two-handed sign language gestures.
Journal Article
Perceiving fingerspelling via point-light displays: The stimulus and the perceiver both matter
Signed languages such as American Sign Language (ASL) rely on visuospatial information that combines hand and bodily movements, facial expressions, and fingerspelling. Signers communicate in a wide array of sub-optimal environments, such as in dim lighting or from a distance. While fingerspelling is a common and essential part of signed languages, the perception of fingerspelling in difficult visual environments is not well understood. The movement and spatial patterns of ASL are well-suited to representation by dynamic Point Light Display (PLD) stimuli in which human movement is shown as an array of moving dots affixed to joints on the body. We created PLD videos of fingerspelled location names. The location names were either Real (e.g., KUWAIT) or Pseudo-names (e.g., CLARTAND), and the PLDs showed either a High or a Low number of markers. In an online study, Deaf and Hearing ASL users (total N = 283) watched 27 PLD stimulus videos that varied by Word Type and Number of Markers. Participants watched the videos and typed the names they saw, along with how confident they were in their response. We predicted that when signers see ASL fingerspelling PLDs, language experience in ASL will be positively correlated with accuracy and self-rated confidence scores. We also predicted that Real location names would be understood better than Pseudo names. Our findings supported those predictions. We also discovered a significant interaction between Age and Word Type, which suggests that as people age, they use outside world knowledge to inform their fingerspelling success. Finally, we examined the accuracy and confidence in fingerspelling perception in early ASL users. Studying the relationship between language experience with PLD fingerspelling perception allows us to explore how hearing status, ASL fluency levels, and age of language acquisition affect the core abilities of understanding fingerspelling.
Journal Article
TFS Point-on-Hand Sign Recognition Using Part Affinity Fields
Our study investigates an application of a bottom-up design for keypoint regression, Part Affinity Fields (PAFs), for sign language recognition. Automatic sign language recognition could facilitate communication between deaf people and the hearing majority. Sign languages generally employ both semantic and finger-spelling signing. Semantic signing includes acting out to convey meaning, while finger spelling complements signing through the spelling out of proper names. Specifically, this article addresses an automatic recognition framework for the static point-on-hand (PoH) signing of Thai Finger Spelling (TFS)—the finger-spelling part of Thai Sign Language (TSL). From a pattern recognition perspective, PoH signing is quite distinct among signing schemes for requirement of precise localization of key parts on the signing hands. A recent study addressed PoH using an off-the-shelf version of MediaPipe Hands (MPH) and found shortcomings particularly when there was a high degree of hand-to-hand interaction. The top-down design of MPH was hypothesized to be the culprit. Our study investigates a bottom-up design, Part Affinity Fields (PAFs), along with examination of the related factors. The results support the hypothesis of a high-degree of hand-to-hand interaction posited by the MPH study. However, the overall performance of the PAF-based approach is shown to be modestly effective (72% accuracy vs. 58% and 47% of the MPH- and X-Pose-based approaches). In addition, its generalization is shown to be lacking. Thus TFS point-on-hand sign recognition remains a challenge.
Journal Article
Variability in the Representation of the ASL Fingerspelled Alphabet
The American Sign Language (ASL) fingerspelled alphabet is often a starting point for novice sign learners. The twenty-six handshapes of the alphabet are typically compiled into visual pedagogical materials designed to help learners master this cornerstone of sign learning. Second-language sign learners often make mistakes in their signing that are related to the fact that signs are visual symbols which thus appear differently depending on one’s perspective. In this study, we analyzed fifty-two commonly available representations of the ASL alphabet to determine the degree of variability exhibited among these materials for general characteristics, such as the medium employed (photographs, digital illustrations, or hand drawings), inclusion of alphabet graphemes and/or object images, and representations of diversity, as well as five parameters related to perspective-taking: perspective on the sign (signer/addressee), angle of hand (0, 45, or 90 degrees), directionality of hand (facing left, right, or front), hand selection (left or right hand), and depiction of movement. We discovered a high degree of variability in the way that ASL handshapes are represented pictorially, with most of the letters of the alphabet exhibiting either moderate or high variability in the perspectives, angles, and directionalities of the hand portrayed. We conclude that there is a great deal of heterogeneity in the way that the ASL finger-spelling alphabet is represented in didactic materials, and we suggest ways that educators and publishers can improve their teaching materials by incorporating multiple visual perspectives.
Journal Article
Semantic processing of iconic signs is not automatic: Neural evidence from hearing non-signers
Iconicity facilitates learning signs, but it is unknown whether recognition of meaning from the sign form occurs automatically. We recorded ERPs to highly iconic (transparent) and non-iconic ASL signs presented to one group who knew they would be taught signs (learners) and another group with no such expectations (non-learners). Participants watched sign videos and detected an occasional grooming gesture (no semantic processing required). Before sign onset, learners showed a greater frontal negativity compared to non-learners for both sign types, possibly due to greater motivation to attend to signs. During the N400 window, learners showed greater negativity to iconic than non-iconic signs, indicating more semantic processing for iconic signs. The non-learners showed a later and much weaker iconicity effect. The groups did not differ in task performance or in P3 amplitude. We conclude that comprehending the form-meaning mapping of highly iconic signs is not automatic and requires motivation and attention.
Journal Article
A Sociolinguistic Analysis of Name Signs in Israeli Sign Language
Name sign systems have been described in many deaf communities around the world. The most frequent name sign types are associated with an individual's appearance, for example, a signers' hairstyle, clothes, and physical features such as height, weight, etc. However, a recent study that examined name signs in Swedish Sign Language, for example, found a decrease in name signs based on appearance and an increase in person name signs, suggesting that name signs are undergoing changes. This study examines name signs produced by 160 deaf signers of Israeli Sign Language (ISL), a sign language that emerged in Israel around ninety years ago. The findings show that, like in other studies, name signs based on appearance are the most frequent in ISL. However, the distribution of name sign types differed based on signers' age and language background. Older signers and deaf people from hearing families are more likely to have name signs related to their appearance while younger signers and deaf people from deaf families are more likely to have name signs related to their legal name, including initialized name signs or signs based on the literal translation of the name. The results are discussed in light of changes in society including changes in deaf education and a rise in political correctness.
Journal Article
The 'Experiment' Explained
John B. Burnet responds to J.A. Jacobs' article \"An important distinction: Methodical signs\" (reprinted in this issue from the following original: Jacobs, J. A. (1853, October). An important distinction: Methodical signs. American Annals of the Deaf and Dumb, 6(1), 51–56.
Journal Article