Method and apparatus for processing video data

A method comprising:

• receiving a plurality of digital images, each image including object pixels for a dynamic target object, non-object pixels for a static object, and non-object pixels for an additional dynamic object;
• aligning a first sequence of digital images with a second sequence of digital images that is not adjacent to the first sequence of digital images, the first and second sequences of digital images included in the plurality of digital images;
• selecting a first digital image from the first sequence of digital images and selecting a second digital image from the second sequence of digital images;
• filtering the non-object pixels for the additional dynamic object of the first digital image from the object pixels for the dynamic target object of the first digital image based on the non-object pixels for the additional dynamic object in the first and second digital images being less similar than the object pixels for the dynamic target object in the first and second digital images;
• selecting a third digital image that includes non-object pixels;
• and
• filtering the non-object pixels for the static object of the first digital image from the object pixels for the dynamic target object of the first digital image based on the non-object pixels in the third digital image being more similar to the non-object pixels for the static object in the first and second digital images than to the object pixels for the dynamic target object in the first and second digital images;
• wherein the dynamic target object and the additional dynamic object are both non-stationary and change locations in adjacent neighboring frames, the dynamic target object is not found in a sequence of frames that is subsequent to a sequence of frames including the dynamic target object, and the third digital image is included in a frame that is non-adjacent and does not neighbor a frame that includes one of the first and second digital images.