基于多尺度卷积注意力和双分支对抗训练的抗压缩鲁棒视频水印OA

To overcome the limitations of current deep learning-based video watermarking methods,such as reliance on single-scale feature extraction,limited adversarial training mechanisms,and insufficient robustness against compression,this paper proposes a robust video watermarking model called MSCA-GAN(multi-scale convolutional attention generative adversarial network),which integrates a multi-scale convolutional attention mechanism and a dual-branch adversarial training framework.The model employs a lightweight multi-scale attention module to extract key features form video frames at both local and global perspectives.Combined with depthwise separable convolution,it reduces computational complexity while achieving precise localization and strength control of watermark embedding,thereby enhancing invisibility.This paper innovatively designs a dual-branch adversarial training structure,in which a learnable adversary network is introduced to simulate real-world attacks,enhancing the model's robustness against common threats such as compression,cropping,and scaling.Experimental results demonstrate that the watermarked videos generated by MSCA-GAN achieve an average PSNR of 44.61 dB and a SSIM of 0.964,significantly outperforming existing methods.Under H.264 compression,the average decoding accuracy reaches 94.01％.Moreover,the model maintains strong robustness even under severe cropping and scaling attacks.In summary,MSCA-GAN provides an efficient and reliable solution for multimedia content copyright protection.It has the potential to be extended to emerging coding standards such as H.265,further enhancing its robustness in complex application scenarios.