DeceptionNet: Network-Driven Domain Randomization (Supplement)

DeceptionNet D: Using the defined nomenclature, the encoding part of the DeceptionNet can be described as: C 64−MP −C 128−C 128−MP −C 128−C 128; and its decoding part as: UP64−C 64−C 64−UP64−C 64−C 64−MB . Where MB is defined by the specific module type, and MP stands for a 2-factor max-pooling layer. Encoding blocks have skip connections concatenating the channels with the opposite decoding blocks. The visual representation of the DeceptionNet’s architecture is depicted in Fig. 4.