End‑to‑end robust video watermarking against screen‑recapturing attacks based on spatio-temporal feature enhancement network

doi:10.11772/j.issn.1001-9081.2025080998

Journal of Computer Applications ›› 2026, Vol. 46 ›› Issue (9): 2889-2897.DOI: 10.11772/j.issn.1001-9081.2025080998

• Cyber security • Previous Articles

End‑to‑end robust video watermarking against screen‑recapturing attacks based on spatio-temporal feature enhancement network

Yimin ZHOU¹^,², Wenhui FU¹^,², Jie LUO¹^,²(), Juan WANG¹^,²

^1.School of Cybersecurity （Shuangliu Industry College），Chengdu University of Information Technology，Chengdu Sichuan 610225，China
^2.SUGON Industrial Control and Security Center （Chengdu University of Information Technology），Chengdu Sichuan 610225，China

Received:2025-09-04 Revised:2025-10-30 Accepted:2025-11-13 Online:2025-12-10 Published:2026-09-10
Contact: Jie LUO
About author:ZHOU Yimin， born in 1980， Ph. D.， professor. His research interests include multimedia big data， internet of vehicles security.
FU Wenhui， born in 2000， M. S. candidate. Her research interests include information hiding， generative artificial intelligence.
LUO Jie， born in 1993， Ph. D.， lecturer. Her research interests include multimedia security， artificial intelligence security.
WANG Juan， born in 1981， Ph. D.， professor. Her research interests include internet of things security， artificial intelligence.
Supported by:
Sichuan Provincial Natural Science Foundation for Distinguished Young Scholars(2025NSFJQ0053);Open Funding of SUGON Industrial Control and Security Center(CUIT-SICSC-2025-01);General Program of Inner Mongolia Natural Science Foundation(2026MS0300)

基于时空特征增强网络的端到端抗屏摄攻击鲁棒视频水印

周益民¹^,², 符雯惠¹^,², 罗杰¹^,²(), 王娟¹^,²

^1.成都信息工程大学网络空间安全学院（双流新型产业学院），成都 610225
^2.先进微处理器国家工程实验室（共建）（成都信息工程大学），成都 610225

通讯作者: 罗杰
作者简介:周益民（1980—），男，四川邛崃人，教授，博士，CCF会员，主要研究方向：多媒体大数据、车联网安全
符雯惠（2000—），女，四川眉山人，硕士研究生，主要研究方向：信息隐藏、生成式人工智能
罗杰（1993—），女，四川南充人，讲师，博士，CCF会员，主要研究方向：多媒体安全、人工智能安全
王娟（1981—），女，四川成都人，教授，博士，CCF会员，主要研究方向：物联网安全、人工智能。
基金资助:
四川省自然科学基金杰出青年科学基金资助项目(2025NSFJQ0053);先进微处理器国家工程实验室（共建）开放课题(CUIT-SICSC-2025-01);内蒙古自然科学基金面上项目(2026MS0300)

Abstract

Abstract:

Recently， with the advancement of multi-source capturing devices and sensing technologies， screen recapturing has become a primary channel for video information leakage during dissemination. The screen-recapturing process causes irreversible physical damage to video watermark information， making it difficult to trace the source of information leakage， which poses a significant challenge in the field of video forensics. The existing anti-screen-recapturing video watermarking techniques are still in the preliminary research stage， and are mainly relying on handcrafted frequency-domain features， which struggle to resist cross-device screen-recapturing attacks effectively. Therefore， an end-to-end anti-screen-recapturing video watermarking method was proposed on the basis of a spatio-temporal feature enhancement network. First， to enrich watermark embedding features， a spatio-temporal feature enhancement network module was designed by integrating an optical flow estimation network and a high-pass filter， thereby optimizing the watermark information embedding strategy. Then， by introducing key factors of screen-recapturing channel noise to simulate screen-recapturing distortion， an anti-screen-recapturing attack noise layer was designed to enhance cross-device robustness against such attacks. Experimental results in real-world screen-recapturing attack scenarios demonstrate that the proposed method achieves a decoding accuracy of over 95%， with both robustness and image quality outperforming the mainstream solutions such as Robust Inherent Video Information Embedding （RIVIE） and Automatic， Robust and Blind Video waterMarking resisting camera recording （ARB-VM）.

Key words: video forensics, video watermarking, screen-recapturing attack, deep learning, robustness

摘要：

近年来，随着多源拍摄设备及传感技术的发展，屏幕翻拍已成为视频信息泄露的主要途径。屏摄过程会对视频水印信息造成不可逆的物理损害，从而难以追踪溯源信息泄露途径，是当前视频取证领域面临的重要挑战。现有抗屏摄视频水印技术尚处于初步研究阶段，主要依赖于手工提取频域特征，难以有效抵御跨设备屏摄攻击。因此，提出一种基于时空特征增强网络的端到端抗屏摄攻击视频水印方法。首先，为丰富水印嵌入特征，结合光流估计网络和高通滤波器设计时空特征增强网络模块，以优化水印信息的嵌入策略；其次，引入屏摄信道噪声关键因素模拟屏摄失真，设计抗屏摄攻击噪声层，增强抗屏摄攻击的跨设备鲁棒性。在真实屏摄攻击场景中的实验结果表明，所提方法的解码准确率超过95%，且它的鲁棒性与图像质量均优于主流的鲁棒固有视频信息隐藏（RIVIE）和自动鲁棒盲视频水印（ARB-VM）方案。

关键词: 视频取证, 视频水印, 屏摄攻击, 深度学习, 鲁棒性

CLC Number:

TP391.41

Yimin ZHOU, Wenhui FU, Jie LUO, Juan WANG. End‑to‑end robust video watermarking against screen‑recapturing attacks based on spatio-temporal feature enhancement network[J]. Journal of Computer Applications, 2026, 46(9): 2889-2897.

周益民, 符雯惠, 罗杰, 王娟. 基于时空特征增强网络的端到端抗屏摄攻击鲁棒视频水印[J]. 《计算机应用》唯一官方网站, 2026, 46(9): 2889-2897.

Figures/Tables 12

Fig. 1 Framework of anti-screen-recapturing video watermarking method

Fig. 2 STFBlock structure

Fig. 3 Operating effects of screen-recapturing noise layer

Tab. 1 Comparison of PSNR and SSIM among different methods?

方法	PSNR/dB	SSIM
DVMark	37.00	0.985
RIVIE	38.73	0.987
本文方法	39.08	0.992

Fig. 4 Visual quality of watermarked video frames?

Fig. 5 Examples of screen-recaptured video frames at different capturing distances

Tab. 2 Comparison of extraction accuracy at different recording distances

距离/cm	不同方法的提取准确率/%
距离/cm	RIVIE	ARB-VM	本文方法
20	97.66	98.89	99.05
30	98.60	98.14	98.75
40	97.80	95.40	96.82
50	96.30	94.79	97.27
60	94.95	95.14	95.59

Fig. 6 Examples of screen-recaptured video frames at different capturing angles?

Tab. 3 Comparison of extraction accuracy rates at different capturing angles

录制角度/（°）	不同方法的提取准确率/%
录制角度/（°）	RIVIE	ARB-VM	本文方法
-45	95.71	94.30	96.67
-30	97.43	93.65	97.62
-15	98.70	95.40	98.65
0	98.60	97.14	98.75
+15	98.50	96.19	98.36
+30	96.39	94.79	97.60
+45	95.30	94.26	95.92

Tab. 4 Comparison of extraction accuracy captured by different devices

录制设备	不同播放设备的提取准确率
录制设备	KTC H27T22C	AOC 24B1
Redmi K80	97.96	96.67
iPhone 13	99.05	98.15
iPhone 16 Pro Max	99.28	98.85
OPPO Find X8	98.23	97.97

Fig. 7 Examples of video frames captured by different devices

Tab. 5 Extraction accuracies and PSNRs under different embedded capacities

载荷/bit	准确率/%	PSNR/dB
64	98.87	39.12
80	98.82	39.11
96	98.75	39.08
112	98.43	38.95

References 29

[1]	Langelaar G C， Setyawan I， Lagendijk R L. Watermarking digital image and video data. A state-of-the-art overview ［J］. IEEE Signal Processing Magazine， 2000， 17（5）： 20-46.
[2]	Evsutin O， Dzhanashia K. Watermarking schemes for digital images： robustness overview ［J］. Signal Processing： Image Communication， 2022， 100： No.116523.
[3]	王翌妃，周杨铭，钱振兴，等. 鲁棒视频水印研究进展［J］. 中国图象图形学报， 2022， 27（1）： 27-42.
	Wang Yifei， Zhou Yangming， Qian Zhenxing， et al. Review of robust video watermarking ［J］. Journal of Image and Graphics， 2022， 27（1）： 27-42.
[4]	Chang Q， Huang L， Liu S， et al. Blind robust video watermarking based on adaptive region selection and channel reference ［C］// MM 2022. New York： ACM， 2022： 2344-2350.
[5]	Wu D， Zhang X， Wang J， et al. Novel robust video watermarking scheme based on concentric ring subband and visual cryptography with piecewise linear chaotic mapping ［J］. IEEE Transactions on Circuits and Systems for Video Technology， 2024， 34（10）： 10281-10298.
[6]	Huang I C， Wu J Y， Ooi W T. RBMark： robust and blind video watermark in DT CWT domain ［J］. Journal of Visual Communication and Image Representation， 2025， 109： No.104438.
[7]	Zhu J， Kaplan R， Johnson J， et al. HiDDeN： hiding data with deep networks ［C］// ECCV 2018， LNCS 11219. Cham： Springer， 2018： 682-697.
[8]	Ahmadi M， Norouzi A， Karimi N， et al. ReDMark： framework for residual diffusion watermarking based on deep networks ［J］. Expert Systems with Applications， 2020， 146： No.113157.
[9]	Fang H， Jia Z， Ma Z， et al. PIMoG： an effective screen-shooting noise-layer simulation for deep-learning-based watermarking network［C］// MM 2022. New York： ACM， 2022： 2267-2275.
[10]	Li P， Pei Y， Li J. A comprehensive survey on design and application of autoencoder in deep learning ［J］. Applied Soft Computing， 2023， 138： No.110176.
[11]	Fang H， Zhang W， Zhou H， et al. Screen-shooting resilient watermarking ［J］. IEEE Transactions on Information Forensics and Security， 2019， 14（6）： 1403-1418.
[12]	Celik M U， Sharma G， Tekalp A M， et al. Lossless generalized-LSB data embedding ［J］. IEEE Transactions on Image Processing， 2005， 14（2）： 253-266.
[13]	Deshmukh P R， Rahangdale B. Hash based least significant bit technique for video steganography ［J］. International Journal of Engineering Research and Applications， 2014， 4（1）： 44-49.
[14]	Chen S， Malik A， Zhang X， et al. A fast method for robust video watermarking based on Zernike moments ［J］. IEEE Transactions on Circuits and Systems for Video Technology， 2023， 33（12）： 7342-7353.
[15]	Yang L， Wang H， Zhang F， et al. Robust video watermarking based on residual synchronization in the DCT domain ［C］// MMSP 2022. Piscataway： IEEE， 2022： 1-6.
[16]	Sun X C， Lu Z M， Wang Z， et al. A geometrically robust multi-bit video watermarking algorithm based on 2-D DFT ［J］. Multimedia Tools and Applications， 2021， 80（9）： 13491-13511.
[17]	Takale S， Mulani A. DWT-PCA based video watermarking ［J］. Journal of Electronics， Computer Networking and Applied Mathematics， 2022， 2（6）： 1-7.
[18]	Huan W， Li S， Qian Z， et al. Exploring stable coefficients on joint sub-bands for robust video watermarking in DT CWT domain［J］. IEEE Transactions on Circuits and Systems for Video Technology， 2022， 32（4）： 1955-1965.
[19]	Kumaraswamy E， Vatti R， Vallathan G， et al. SVD based robust unsighted video watermarking technique for different attacks ［J］. IOP Conference Series： Materials Science and Engineering， 2020， 981（3）： No.032030.
[20]	Lin L， Wu D， Wang J， et al. Automatic， robust and blind video watermarking resisting camera recording ［J］. IEEE Transactions on Circuits and Systems for Video Technology， 2024， 36（12）： 13413-13426.
[21]	罗志伟，刘持标. 基于H.266/通用视频编码的帧内双模式的视频水印算法［J］. 计算机应用， 2021， 41（2）： 456-460.
	Luo Zhiwei， Liu Chibiao. A Video Watermarking algorithm based on H.266/versatile video coding with intra frame dual mode ［J］. Journal of Computer Applications， 2021， 41（2）： 456-460.
[22]	蔡春亭，冯桂，王驰，等. 基于帧内预测模式多划分的HEVC鲁棒视频水印算法［J］. 计算机应用， 2017， 37（6）： 1772-1776.
	Cai Chunting， Feng Gui， Wang Chi， et al. Robust video watermarking algorithm for HEVC based on intra-frame prediction modes of multi-partitioning ［J］. Journal of Computer Applications， 2017， 37（6）： 1772-1776.
[23]	Wang Y， Pearmain A. Blind MPEG-2 video watermarking robust against geometric attacks： a set of approaches in DCT domain ［J］. IEEE Transactions on Image Processing， 2006， 15（6）： 1536-1543.
[24]	Mishra A， Bansal M， Sharma A. Video watermarking of live streamed MPEG-4 frames using ELM-Fuzzy-PSO hybrid scheme［J］. Multimedia Tools and Applications， 2024， 83（14）： 41997-42035.
[25]	Ullah R， Khan S D， Ullah M， et al. Toward authentication of videos： integer transform based motion vector watermarking ［J］. IEEE Access， 2022， 10： 75063-75073.
[26]	Li Z， Jiang X， Dong Y， et al. An anti-steganalysis HEVC video steganography with high performance based on CNN and PU partition modes ［J］. IEEE Transactions on Dependable and Secure Computing， 2023， 20（1）： 606-619.
[27]	李琦，王春鹏，王晓雨，等. 基于残差学习的新型不可感知水印攻击方法［J］. 软件学报， 2023， 34（9）： 4351-4361.
	Li Qi， Wang Chunpeng， Wang Xiaoyu， et al. Novel imperceptible watermark attack method based on residual learning ［J］. Journal of Software， 2023， 34（9）： 4351-4361.
[28]	Luo X， Li Y， Chang H， et al. DVMark： a deep multiscale framework for video watermarking ［J］. IEEE Transactions on Image Processing， 2025， 34： 4371-4385.
[29]	Jia J， Gao Z， Zhu D， et al. RIVIE： robust inherent video information embedding ［J］. IEEE Transactions on Multimedia， 2023， 25： 7364-7377.

End‑to‑end robust video watermarking against screen‑recapturing attacks based on spatio-temporal feature enhancement network

基于时空特征增强网络的端到端抗屏摄攻击鲁棒视频水印

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 12

References 29

Related Articles 15

Recommended Articles

Metrics

[1]	Wei CUI, Ping ZHANG, Zhuoling XIAO, Zhuohang CHEN. Incremental semantic mapping system based on visual SLAM [J]. Journal of Computer Applications, 2026, 46(9): 2800-2808.
[2]	Shang LIU, Zhaosen TANG, Hongyue LIU, Linfang DONG, Jin ZHOU. Multimodal sentiment analysis model for missing modalities under shared semantic conditions [J]. Journal of Computer Applications, 2026, 46(9): 2761-2768.
[3]	Ming LIU, Dongqi SHEN, Ziyang MENG. Point cloud registration network with dual-branch multi-level feature fusion [J]. Journal of Computer Applications, 2026, 46(8): 2584-2593.
[4]	Fengchun LIU, Xinying SHAO, Chunying ZHANG, Liya WANG, Jing REN. FCMdepth： monocular depth estimation framework with multi-scale feature optimization [J]. Journal of Computer Applications, 2026, 46(8): 2603-2611.
[5]	Wei LIU, Weigang LI, Zhiqiang TIAN. Representation learning method of hierarchical rotation-invariant geometric structure for point cloud classification and segmentation [J]. Journal of Computer Applications, 2026, 46(8): 2594-2602.
[6]	Xiaojin GUO, Xuyang SUI, Kenan ZHOU. Channel estimation of STAR-RIS-assisted hybrid-field communication system based on deep learning [J]. Journal of Computer Applications, 2026, 46(8): 2548-2554.
[7]	Xinliang LIU, Yushi XU, Dubai LI, Yanzhao REN. Survey of knowledge graph-based question answering methods [J]. Journal of Computer Applications, 2026, 46(8): 2394-2410.
[8]	Jing LIU, Shaoze ZHAO, Xingang LIU, Haozhe NIU, Haipeng JI. Cross-condition microstructure data generation method for titanium alloys based on improved CGAN [J]. Journal of Computer Applications, 2026, 46(7): 2373-2382.
[9]	Miaogen LING, Rui JING, Wei FANG. Survey of research on applications of explainable deep learning in tropical cyclone forecasting [J]. Journal of Computer Applications, 2026, 46(7): 2318-2326.
[10]	Xiangyi WU, Hailiang YE, Feilong CAO. Point cloud completion method based on smooth-sharpen graph convolution [J]. Journal of Computer Applications, 2026, 46(7): 2267-2276.
[11]	Xiuli DU, Xing GAO, Xiaoyu ZHANG, Chengsheng PAN, Qijie ZOU. Video snapshot compressive imaging reconstruction method based on dense spatio-temporal deformable attention [J]. Journal of Computer Applications, 2026, 46(7): 2288-2296.
[12]	Xue XU, Hu FAN, Yandan WANG, Xue DING, Xuefeng GAO, Bo ZHANG, Bo LIU, Beihong JIN. Multi-view consistency-driven robust feature selection method [J]. Journal of Computer Applications, 2026, 46(6): 1844-1854.
[13]	Yi DU, Mingjin XU, Jiayi KONG, Liyao WANG, Chen ZHAO. Low-rank adaptive parameter-efficient fine-tuning algorithm based on YOLOv11 [J]. Journal of Computer Applications, 2026, 46(6): 1738-1745.
[14]	Zhenkai XIONG, Mengjun XU, Yinyin SUN, Xin WANG. Maritime ship detection algorithm under complex weather environments based on enhanced YOLOv8 [J]. Journal of Computer Applications, 2026, 46(6): 1998-2006.
[15]	Lili HE, Meng CAO, Lei ZHANG, Hongjun PAN, Yi LIU, Chengxin SUN. Sign language generation model based on Kolmogorov-Arnold network and diffusion Transformer [J]. Journal of Computer Applications, 2026, 46(6): 1801-1810.