scholarly journals An Attention Self-Supervised Contrastive Learning Based Three-Stage Model for Hand Shape Feature Representation in Cued Speech

2021 ◽  
Author(s):  
Jianrong Wang ◽  
Nan Gu ◽  
Mei Yu ◽  
Xuewei Li ◽  
Qiang Fang ◽  
...  
Keyword(s):  
Feature Representation ◽  
Stage Model ◽  
Shape Feature ◽  
Hand Shape ◽  
Cued Speech

Phonological Similarity Effects in Memory for Serial Order of Cued Speech

2001 ◽  
Vol 44 (5) ◽  
pp. 949-963 ◽  
Author(s):  
Jacqueline Leybaert ◽  
Josiane Lechat
Keyword(s):  
Visual Stimuli ◽  
Serial Order ◽  
Phonological Similarity ◽  
Recency Effect ◽  
The Other ◽  
Auditory Stimuli ◽  
Linguistic Experience ◽  
Lower Performance ◽  
Hand Shape ◽  
Cued Speech

Experiment I investigated memory for serial order by congenitally, profoundly deaf individuals, 6–22 years old, for words presented via Cued Speech (CS) without sound. CS is a system that resolves the ambiguity inherent in speechreading through the addition of manual cues. The phonological components of CS are mouth shape, hand shape, and hand placement. Of interest was whether the recall of serial order was lower for lists of words similar in both mouth shape and hand placement, or similar in mouth shape only, or in hand placement only than for control lists designed to minimize these similarities. Deaf participants showed lower performance on the three similar lists than the control lists, suggesting that deaf individuals use the phonology of CS to support their recall. In Experiment II, the same lists were administered to two groups of hearing participants. One group, experienced producers of CS, received the CS stimuli without sound; the other group, unfamiliar with CS, received the CS stimuli audiovisually. Participants experienced with CS showed no effect of hand placement similarity, suggesting that this effect may be related to the linguistic experience of deaf participants. The recency effect was greater in the hearing group provided with sound, indicating that the traces left by auditory stimuli are perceptually more salient than those left by the visual stimuli encountered in CS.

scholarly journals Robust Hand Shape Features for Dynamic Hand Gesture Recognition Using Multi-Level Feature LSTM

2020 ◽  
Vol 10 (18) ◽  
pp. 6293
Author(s):  
Nhu-Tai Do ◽  
Soo-Hyung Kim ◽  
Hyung-Jeong Yang ◽  
Guee-Sang Lee
Keyword(s):  
Gesture Recognition ◽  
Point Cloud ◽  
Hand Gesture Recognition ◽  
Feature Representation ◽  
Geometric Transformation ◽  
Hand Gesture ◽  
Shape Features ◽  
Hand Shape ◽  
Dynamic Hand Gesture Recognition ◽  
Multi Level

This study builds robust hand shape features from the two modalities of depth and skeletal data for the dynamic hand gesture recognition problem. For the hand skeleton shape approach, we use the movement, the rotations of the hand joints with respect to their neighbors, and the skeletal point-cloud to learn the 3D geometric transformation. For the hand depth shape approach, we use the feature representation from the hand component segmentation model. Finally, we propose a multi-level feature LSTM with Conv1D, the Conv2D pyramid, and the LSTM block to deal with the diversity of hand features. Therefore, we propose a novel method by exploiting robust skeletal point-cloud features from skeletal data, as well as depth shape features from the hand component segmentation model in order for the multi-level feature LSTM model to benefit from both. Our proposed method achieves the best result on the Dynamic Hand Gesture Recognition (DHG) dataset with 14 and 28 classes for both depth and skeletal data with accuracies of 96.07% and 94.40%, respectively.

scholarly journals A two-stage model of V2 demonstrates efficient higher-order feature representation

2021 ◽  
Vol 21 (9) ◽  
pp. 2654
Author(s):  
Timothy D. Oleskiw ◽  
Ruben R. Diaz-Pacheco ◽  
J. Anthony Movshon ◽  
Eero P. Simoncelli
Keyword(s):  
Higher Order ◽  
Feature Representation ◽  
Stage Model ◽  
Two Stage

The Solution Strategy as an Indicator of the Developmental Stage of Preschool Children’s Mental-Rotation Ability

2010 ◽  
Vol 31 (2) ◽  
pp. 95-100 ◽  
Author(s):  
Claudia Quaiser-Pohl ◽  
Anna M. Rohe ◽  
Tobias Amberger
Keyword(s):  
Preschool Children ◽  
Mental Rotation ◽  
Latent Class ◽  
Preschool Age ◽  
Stage Model ◽  
Solution Strategy ◽  
Thinking Aloud ◽  
Transition Analysis ◽  
Solution Strategies ◽  
Close Relationship

The solution strategies of preschool children solving mental-rotation tasks were analyzed in two studies. In the first study n = 111 preschool children had to demonstrate their solution strategy in the Picture Rotation Test (PRT) items by thinking aloud; seven different strategies were identified. In the second study these strategies were confirmed by latent class analysis (LCA) with the PRT data of n = 565 preschool children. In addition, a close relationship was found between the solution strategy and children’s age. Results point to a stage model for the development of mental-rotation ability as measured by the PRT, going from inappropriate strategies like guessing or comparing details, to semiappropriate approaches like choosing the stimulus with the smallest angle discrepancy, to a holistic or analytic strategy. A latent transition analysis (LTA) revealed that the ability to mentally rotate objects can be influenced by training in the preschool age.

Sign in / Sign up

Export Citation Format

Share Document