Carrying out AI and object recognition to form recyclables is complicated and would require an embedded chip capable of dealing with these features with significant efficiency.  We represent video clips and images as collections of smaller sized models of information known as patches, Ev