Contactless Air Canvas Using MediaPipe and LSTM for Temporal Jitter Reduction.

This paper presented an enhancement to our previous OpenCV- and MediaPipe-based Air Canvas system by introducing an LSTM-based temporal filter to address the positional jitter inherent in frame-by-frame hand-landmark detection. The proposed method reduced the positional error from ±7 px to ±2 px while improving real-time performance to up to 30 FPS, demonstrating that contactless drawing can be made practically viable. Beyond this technical improvement, the system offers several real-world benefits: it provides an accessible, marker-free drawing interface for individuals with mobility impairments, consistent with prior findings on touchless gesture interfaces as assistive technology [20]; it supports paperless creative and educational practices that reduce dependence on physical drawing materials; and it gives artists a flexible digital canvas for creative expression. These results confirm that learning-based temporal filtering is an effective and lightweight solution for improving the usability of vision-based gesture interfaces.

1. F. Zhang, V. Bazarevsky, A. Vakunov, A. Tkachenka, G. Sung, C.-L. Chang, and M. Grundmann, “MediaPipe Hands: On-device real-time hand tracking,” arXiv:2006.10214, 2020. 2.   S. Hochreiter and J. Schmidhuber, “Long short-term memory,” Neural Computation, vol. 9, no. 8, pp. 1735–1780, 1997. 3.   P. C. Patil and N. Chaudhari, “Review of contactless drawing interfaces,” Int. J. of Research, 2025. 4.   G. Bradski, “The OpenCV library,” Dr. Dobb’s Journal of Software Tools, vol. 25, no. 11, pp. 120–126, 2000. 5.   S. S. Rautaray and A. Agrawal, “Vision based hand gesture recognition for human computer interaction: a survey,” Artificial Intelligence Review, vol. 43, no. 1, pp. 1–54, 2015. 6.   G. Welch and G. Bishop, “An introduction to the Kalman filter,” Tech. Rep. TR 95-041, Dept. of Computer Science, Univ. of North Carolina at Chapel Hill, 1995. 7.   G. Casiez, N. Roussel, and D. Vogel, “1€ filter: A simple speed-based low-pass filter for noisy input in interactive systems,” in Proc. ACM CHI Conf. on Human Factors in Computing Systems, 2012, pp. 2527–2530. 8.   M. S. M. Asaari and S. A. Suandi, “Hand gesture tracking system using Adaptive Kalman Filter,” in Proc. 10th Int. Conf. on Intelligent Systems Design and Applications (ISDA), Cairo, Egypt, 2010, pp. 166–171. 9. P. Xu, “A real-time hand gesture recognition and human-computer interaction system,” arXiv:1704.07296, 2017. 10. C. Cui, M. S. Sunar, and G. E. Su, “Deep vision-based real-time hand gesture recognition: a review,” PeerJ Computer Science, 2025.