• Prafull Pradeep Sonawane
• Aniket Arun Shinde
• Linesh Pradip Gahival
• Prof.S.V.Mahale

The current usability of eye-typing systems that rely on virtual keyboards is limited due to the lack of adaptive and user-centered approaches. This leads to low text entry rates and the need for frequent recalibration. In this study, we propose a set of methods to address these issues in gaze-based virtual keyboards. Our proposed methods focus on adapting the dwell time in asynchronous mode and the trial period in synchronous mode. We consider commands that allow for corrections within the application. The methods have been tested on a newly developed virtual keyboard designed for a structurally complex language. The keyboard incorporates a two- stage tree-based character selection arrangement. We introduce various mechanisms that are dwell-based or dwell-free, with the added feature of multimodal access. This means that the search for a target item is facilitated through gaze detection, while the selection can be made using dwell time, soft-switch, or gesture detection with surface electromyography in asynchronous mode. In synchronous mode, both the search and selection can be performed solely with the eye-tracker. To evaluate the performance of the system, we conducted experiments under 20 different conditions, measuring text entry rate and information transfer rate. Our adaptive strategy for parameter adaptation over time has demonstrated a significant improvement of over 40% compared to non-adaptive approaches, particularly for new users. The multimodal dwell-free mechanism, which combines eye-tracking and soft-switch, outperforms adaptive methods that rely solely on eye-tracking. Furthermore, we assessed the system's usability using the System Usability Scale and the NASA Task Load Index. The system received an excellent rating on the adjective rating scale, indicating its user-centered focus, while the weighted rating on the NASA Task Load Index was low.

CNN,RNN,KERAS,PYTHON

"Virtual mouse control using Machine learning", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 4, page no.g132-g136, April-2024, Available :http://www.jetir.org/papers/JETIR2404612.pdf

"Virtual mouse control using Machine learning", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 4, page no. ppg132-g136, April-2024, Available at : http://www.jetir.org/papers/JETIR2404612.pdf