This chapter deals with a brief discussion of Two-dimensional (2D) filters for “upsampling” image data. It presents 2D finite impulse response (FIR) filters which are widely used in image and video processing. Almost all image filtering involves FIR rather than infinite impulse response filters because the spatial causality constraints must be relaxed in order for a small 2D FIR convolution filter to process a much larger image by scanning and filtering. The smaller one makes the process noise, the more the measurements are smoothed by the tracking filter, and also, the tracking filter will be very sluggish to track real changes in the target kinematics. The 2D FIR filters are most often used to process image data from camera systems. The process of computing a brightness output based on a weighted average of the adjacent pixels is likely where the term “moving average” for FIR digital filters originates.