A Comparative Study of Deep Learning Models for American SL Recognition
Bibliographic record
Abstract
American SL (ASL) is a natural language employed by the deaf community in the United States as well as in some areas of Canada. ASL is a visual language that makes use of facial expressions, hand signs, and body language as means of communication, effectively conveying meaning to its users. In this research paper, we performed a comparative study of various deep learning approaches with the aim of accurately recognizing hand gestures in ASL. To conduct our comparative study, we created a comprehensive dataset of ASL hand gestures from A to Z, excluding J and Z, and trained each model on this dataset, it also included images from many sources with a range of backgrounds, lighting, and other environmental factors, making the dataset more resilient and adaptable. The models we evaluated included RNN, ConvNeXt, and Vision Transformer, as well as a computer vision-based approach. We assessed the accuracy of each model on each alphabet and the overall accuracy of each model across all alphabets. Our results shows that ConvNeXt achieve the highest overall accuracy of 99.670%, followed by RNN with 96.6664% accuracy and Vision Transformer with 95.2464% accuracy. These findings highlight the importance of using diverse deep-learning models and comprehensive datasets to accurately recognize ASL hand gestures.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.004 | 0.008 |
| Meta-epidemiology (narrow) | 0.002 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.003 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".