Abstract: Systems and methods for representing images with palettes are disclosed. The system may include a processor, a memory communicatively coupled to the processor, and an image to palette representation logic. The image to palette representation logic can receive an image, and identify objects in the image. The image to palette representation logic can further determine colors for each identified object, and identify areas of each object that comprise each color. The image to palette representation logic can also calculate areas comprising each color, and generate a palette based on the calculated overall areas. The image to palette representation logic can further calculate a vector associated with the image based on a vector summation of the overall areas and calculate a vector summation for the generated palette. Finally, the image to palette representation logic can store the palette if a closeness ratio associated with the palette is larger than a threshold.

Abstract: Systems and methods for representing phrases with palettes are disclosed. The system may include a processor, a memory communicatively coupled to the processor, and a logic. The logic can receive a phrase, generate a set of vectors associated with each identified words in the phrase, calculate a first overall vector associated with the received phrase based on a vector summation of the generated set of vectors, generate a palette for a set of colors including a predefined number of colors, and calculate a second overall vector associated with the palette based on a vector summation of the set of colors. The logic can further store the generated palette in response to a determination that a closeness ratio associated with the generated palette is larger than a predetermined threshold. The closeness ratio is defined as an inverse of a difference between the calculated overall vectors.

Abstract: A device to represent an image with a palette is disclosed. The device includes a processor, a memory communicatively coupled to the processor, and a logic. The logic includes a neural network to receive an input data to generate prediction data. The neural network includes a multi-step convolution pathway including a plurality of convolution steps, and a multi-step upsampling pathway including a plurality of upsampling steps. The plurality of upsampling steps includes an input to receive output data from a corresponding convolution pathway step. In response to receiving the input data, feature map output data is generated at the plurality of convolution steps, and at least one step of the plurality of upsampling steps utilizes at least the generated feature map data to generate prediction data. The neural network extracts one or more features from the input data, and generate the prediction data based on the one or more features.