Augmented Semantic Keypoint Set (55, 75, 100) Extracted From Include50 ISL Dataset

Feature extraction was performed using MediaPipe Holistic, capturing a raw total of 543 keypoints per frame across the (x, y, z) axes. Using a Semantic Selection approach, we distilled this data into three specialized configurations—55, 75, and 100 keypoints—tailored for Indian Sign Language recognition. Each configuration includes augmented samples and a standardized splits.json file. This file ensures reproducibility by fixing the data partition at 80% for training, 10% for validation, and 10% for testing. This derivative work is built upon the Include50 dataset; please ensure proper attribution to the original authors (Sridhar et al., 2020) when utilizing these files. Citation @inproceedings{10.1145/3394171.3413528,author = {Sridhar, Advaith and Ganesan, Rohith Gandhi and Kumar, Pratyush and Khapra, Mitesh},title = {INCLUDE: A Large Scale Dataset for Indian Sign Language Recognition},year = {2020},isbn = {9781450379885},publisher = {Association for Computing Machinery},doi = {10.1145/3394171.3413528},numpages = {10},series = {MM '20}}