Hand detection in American Sign Language depth data using domain-driven random forest regression

Zafrulla, Zahoor; Sahni, Himanshu; Bedri, Abdelkareem; Thukral, Pavleen; Starner﻿﻿, Thad

doi:10.1109/fg.2015.7163135

Cited by 7 publications

(6 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recognition experiments with simple body or hand movements can be found in works [22,23]. In [1][2][3][4][5][6][7][24][25], in turn, sign language gestures were recognised using s keletal data. In [26], only depth maps obtained from the sensor were used.…”

Section: Related Workmentioning

confidence: 99%

“…Halim and Abbas in [5] presented DTW -based approach for Pakistani Sign Language recognition with accuracy of 91%. Zafrulla et al in [6] presented American Sign Language recognition dataset and an approach based on random forest regression. The approach used depth images in order to improve Microsoft Kinect Skeleton Tracker.…”

Section: Related Workmentioning

confidence: 99%

“…Gesture recognition technology may help people with hearing disabilit ies to be understood. Sign language recognition, as a challenging examp le of gesture recognition problem, is also the subject of many recent approaches, e.g., [1][2][3][4][5][6][7]. Here, approaches handle large gesture variation, i.e., gesture's executions from a given class can be performed with d ifferent speeds, and there is a dependency of gesture execution on signer's anatomical constraints or intention.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

An Approach to Gesture Recognition with Skeletal Data Using Dynamic Time Warping and Nearest Neighbour Classifier

Ribó¹,

Warchoł²,

Oszust³

2016

IJISA

View full text Add to dashboard Cite

Gestures are natural means of co mmun ication between humans, and therefore their application would benefit to many fields where usage of typical input devices, such as keyboards or joysticks is cu mbersome or unpractical (e.g., in noisy environ ment). Recently, together with emergence of new cameras that allow obtaining not only colour images of observed scene, but also offer the software developer rich informat ion on the number of seen hu mans and, what is most interesting, 3D positions of their body parts, practical applications using body gestures have become more popular. Such informat ion is presented in a form o f skeletal data. In this paper, an approach to gesture recognition bas ed on skeletal data using nearest neighbour classifier with dynamic time warping is presented. Since similar approaches are widely used in the literature, a few practical improvements that led to better recognition results are proposed. The approach is extensively evaluated on three publicly availab le gesture datasets and compared with state-of-the-art classifiers. For some gesture datasets, the proposed approach outperformed its competitors in terms of recognition rate and time of recognition.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

An Approach to Gesture Recognition with Skeletal Data Using Dynamic Time Warping and Nearest Neighbour Classifier

Ribó¹,

Warchoł²,

Oszust³

2016

IJISA

View full text Add to dashboard Cite

show abstract

“…Vision-based approaches usually rely on the videos of sign language hand gestures recorded from a camera (e.g., RGB color cameras [8,15,66], Kinect depth cameras, and time-of-flight (ToF) cameras). Early research used hand-crafted features (e.g., the edge orientation histogram [43], upper body joints [21], skeleton information [71], hand shape features [58,68,70]) from images to build the ASL recognition systems. Zafrulla et al [69] developed an ASL recognition system using multimodal Kinect system.…”

Section: Sign Language Recognitionmentioning

confidence: 99%

SonicASL

Jin

Gao

Zhu

et al. 2021

Proc. ACM Interact. Mob. Wearable Ubiquitous Technol.

View full text Add to dashboard Cite

We propose SonicASL, a real-time gesture recognition system that can recognize sign language gestures on the fly, leveraging front-facing microphones and speakers added to commodity earphones worn by someone facing the person making the gestures. In a user study (N=8), we evaluate the recognition performance of various sign language gestures at both the word and sentence levels. Given 42 frequently used individual words and 30 meaningful sentences, SonicASL can achieve an accuracy of 93.8% and 90.6% for word-level and sentence-level recognition, respectively. The proposed system is tested in two real-world scenarios: indoor (apartment, office, and corridor) and outdoor (sidewalk) environments with pedestrians walking nearby. The results show that our system can provide users with an effective gesture recognition tool with high reliability against environmental factors such as ambient noises and nearby pedestrians.

show abstract

“…The introduction of inexpensive depth cameras has made real‐time hand detection possible with low cost. The use of depth cameras in sign recognition applications has been widely demonstrated [13–15].…”

Section: Introductionmentioning

confidence: 99%