American Sign Language dataset for semantic communications
Description
The dataset offers extensive possibilities for a variety of applications. It can be utilized to study and analyze the positional relationships of joints and fingers in the context of sign language recognition. Moreover, the dataset is a useful tool for training models of artificial intelligence (AI) and machine learning (ML). Using the color-coded characteristics and thorough key-point annotations, researchers may create algorithms for gesture detection, sign language interpretation, and other human-computer interaction chores. RGB photos also help to build models that investigate the advantages of color information in feature extraction and discrimination, therefore enabling possibly more powerful recognition systems. The dataset can also help research on data efficiency since its design shows how to reach high performance measures using small training samples. The dataset consists of 24 folders for training and testing, each corresponding to a specific letter of the American Sign Language (ASL) alphabet. Each folder contains 440 unique images of the same letter for training and 75 for testing.
Resources
| Name |
Format |
Description |
Link |
|
0 |
|
http://data.europa.eu/88u/dataset/oai-zenodo-org-14635573 |
|
0 |
|
http://data.europa.eu/88u/dataset/oai-zenodo-org-14635573 |