[Retracted] Deep Image Watermarking to JPEG Compression Based on Mixed-Frequency Channel Attention

Table 1

Comparison with the SOTA. We realized the model opening source in [28], while directly using the results included as reported in [26, 27] under quality factor 50. However, SSIM is not reported in [26, 27], for which we empty these items. PSNR is measured for RGB channels, except in [26]; they use the Y channel of the YUV color space.

Model	HiDDeN [26]	TSDL [27]	MBRS [28]	Ours
Image size
Message length	30	30	64	64
Noise layer	JPEG-Mask	JPEG	Mixed	Mixed
PSNR	30.09	33.51	36.49	38.13
SSIM	—	—	0.9173	0.9472
BER	15%	22.3%	0.0092%	0.0078%