Kenyan sign language word-based pose dataset
Bibliographic record
Abstract
In an era where technology fosters inclusion, sign language remains underrepresented in linguistic datasets, especially for low-resource languages such as Kenyan Sign Language (KSL). This paper presents a novel dataset created using MediaPipe's pose estimation technology, designed to address the scarcity of resources for KSL. The dataset includes 20,000 video recordings of KSL gestures, converted into anonymized stickman representations alongside detailed 3D pose coordinates stored in .npy files. The data collection process focused on preserving participant privacy while ensuring the integrity of gesture data. By utilizing pose estimation, the dataset captures manual and non-manual features of KSL while maintaining the anonymity of signers. Stickman representations abstract human features, mitigating ethical concerns associated with traditional video datasets and aligning with privacy-preserving practices. The dataset spans a diverse range of themes relevant to KSL, including daily interactions, cultural expressions, and educational contexts, providing comprehensive coverage of the KSL lexicon. This dataset is designed for reuse across multiple domains. Researchers can leverage it to train machine learning models for sign language recognition, while educators can utilize it to develop interactive language learning tools. Additionally, it supports the development of virtual sign language interpreters and 3D avatars for accessibility applications. By enabling seamless integration into machine learning frameworks, the dataset facilitates advancements in KSL-related technologies and contributes to bridging communication gaps within the Deaf community.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 teacher head, 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".