Virtual Paint

Abstract

While most human-machine interactions today take place via direct contact methods like the mouse, keyboard, remote control, touch screen, etc., most human-human interactions take place via non-contact methods like sound and physical movement, which are more natural and intuitive. Many researchers have worked to train machines to recognize human-like nonverbal cues, such as speech, facial expressions, body language, and gestures, in the hopes that they will one day be able to communicate with greater fluidity and efficiency. The most fundamental aspect of human language is gesture, and gestures play crucial functions in human communication. They're the most straightforward approach of conveying information between people and machines. Sign language recognition, robotics, and other fields may all benefit from gesture recognition. Wearable sensor-based approaches and optical camera-based methods are the two simplest ways to classify the devices utilized in gesture recognition. The data glove is an example of a device employed in the wearable sensor-based approach that can accurately capture the motion characteristics of the user's hands, leading to improved recognition accuracy. Wearable sensors are costly to produce and have an adverse effect on the fluidity of user engagement. The use of optical cameras that record a sequence of pictures is at the heart of the optical camera-based technique for distant gesture recording. These visionbased techniques, which use optical cameras to identify motions, work by evaluating visual information collected from pictures. It is challenging to identify and track the hands accurately with optical cameras due to their sensitivity to lighting conditions and clutter backgrounds. Both the optical camera-based method and the wearable sensor-based method have their benefits and drawbacks; however, the project's primary objective was to create a simple and inexpensive approach to enhancing HCI.