Following code was performed in Google Collab and it extracts landmarks from Mediapipe Holistic and train the coordinates for Mexican Sign Language Recognition in its dactylology (29 characters) and the first 10 digits. We adapted the following code to complement our project:
Renotte, N. (2020). ActionDetectionforSignLanguage. GitHub. https://github.com/nicknochnack/ActionDetectionforSignLanguage
TheAIGuysCode. (2020). Colab-webcam. GitHub. https://github.com/theAIGuysCode/colab-webcam
The dataset used were published in Data Mendeley as 7 Open Source datasets, and they can be found here:
Rodriguez, M., Outmane, O., Bassam, A., & Noureddine, L. (2023a). Mexican Sign Language’s Dactylology and Ten First Numbers -Raw videos. From person #1 to #3 [Data set]. Mendeley. V1. https://doi.org/10.17632/48XYBSMVPV.1
Rodriguez, M., Outmane, O., Bassam, A., & Noureddine, L. (2023b). Mexican Sign Language’s Dactylology and Ten First Numbers -Raw videos. From person #4 to #7 [Data set]. Mendeley. V1. https://doi.org/10.17632/JP4YMF2VJW.1
Rodriguez, M., Outmane, O., Bassam, A., & Noureddine, L. (2023c). Mexican Sign Language’s Dactylology and Ten First Numbers -Raw videos. From person #8 to #10 [Data set]. Mendeley. V1. https://doi.org/10.17632/69FMDB25XM.1
Rodriguez, M., Outmane, O., Bassam, A., & Noureddine, L. (2023d). Mexican Sign Language’s Dactylology and Ten First Numbers -Raw videos. Person #11 [Data set]. Mendeley. V1. https://doi.org/10.17632/3DZN5RSTWX.1
Rodriguez, M., Outmane, O., Bassam, A., & Noureddine, L. (2023e). Mexican Sign Language’s Dactylology and Ten First Numbers—Extracted Features and Models [Data set]. Mendeley. V1. https://doi.org/10.17632/HMSC33HMKZ.1
Rodriguez, M., Outmane, O., Bassam, A. & Noureddine, L. (2023f). Mexican Sign Language’s Dactylology and Ten First Numbers—Labeled images and videos. From person #1 to #5 [Data set]. Mendeley. V1. https://doi.org/10.17632/5S4MT7XRD9.1
Rodriguez, M., Outmane, O., Bassam, A., & Noureddine, L. (2023g). Mexican Sign Language’s Dactylology and Ten First Numbers—Labeled images and videos. From person #6 to #11 [Data set]. Mendeley. V1. https://doi.org/10.17632/67HTNZMWBB.1