Abstract: Embodiments are disclosed for a machine learning-based chroma keying process. The method may include receiving an input including an image depicting a chroma key scene and a color value corresponding to a background color of the image. The method may further include generating a preprocessed image by concatenating the image and the color value. The method may further include providing the preprocessed image to a trained neural network. The method may further include generating, using the trained neural network, an alpha matte representation of the image based on the preprocessed image.

Abstract: Disclosed are an infrared touch screen multipoint recognizing method and an infrared touch screen. The method comprises the following steps: driving infrared transmitting tubes and infrared receiving tubes to scan a touch detection area and saving scan data (601); reading scan data of the infrared transmitting tubes and the infrared receiving tubes, to determine boundaries of quasi-touch points and obtain a quasi-touch point set (602); reading scan data of the infrared transmitting tubes and the infrared receiving tubes aligned with a first direction off-axis, to obtain a first calibrated touch point set (603); reading scan data of the infrared transmitting tubes and the infrared receiving tubes aligned with a second direction off-axis, to obtain a second calibrated touch point set (604); and comparing the quasi-touch point set, the first calibrated touch point set and the second calibrated touch point set, to eliminate ghost points and determine real touch points (605).

Abstract: Disclosed are an infrared touch screen multipoint recognizing method and an infrared touch screen. The method comprises the following steps: driving infrared transmitting tubes and infrared receiving tubes to scan a touch detection area and saving scan data (601); reading scan data of the infrared transmitting tubes and the infrared receiving tubes, to determine boundaries of quasi-touch points and obtain a quasi-touch point set (602); reading scan data of the infrared transmitting tubes and the infrared receiving tubes aligned with a first direction off-axis, to obtain a first calibrated touch point set (603); reading scan data of the infrared transmitting tubes and the infrared receiving tubes aligned with a second direction off-axis, to obtain a second calibrated touch point set (604); and comparing the quasi-touch point set, the first calibrated touch point set and the second calibrated touch point set, to eliminate ghost points and determine real touch points (605).