[Retracted] The Effectiveness of Pictorial Aesthetics Based on Multiview Parallel Neural Networks in Art-Oriented Teaching

Table 1

Deep CNN infrastructure.

Number of layers	Kernel size	Feature map size
Conv1-2	3 × 3(1)	224 × 224 × 64
Max-pool l	2 × 2(2)	112 × 112 × 64
Conv3-4	3 × 3(1)	112 × 112 × 128
Max-pool 2	2 × 2(2)	56 × 56 × 128
Conv5-7	3 × 3(1)	56 × 56 × 256
Max-pool 3	2 × 2(2)	28 × 28 × 256
Conv8-10	3 × 3(1)	28 × 28 × 512
Max-pool 4	2 × 2(2)	14 × 14 × 512
Conv11-12	3 × 3(1)	14 × 14 × 512
Conv13 (6 groups)	3 × 3(1)	14 × 14 × 64
fc1		1 × 1 × 512
fc2		1 × 1 × 256
Regression		1 × 1 × 1