Protect, show, attend and tell: Empowering image captioning models with ownership protection

PATTERN RECOGNITION

卷:122

Article

Lim, Jian Han¹  Chan, Chee Seng¹  Ng, Kam Woh²  Fan, Lixin²  Yang, Qiang³ 

[1] Univ Malaya, Fac Comp Sci & Informat Technol, Ctr Image & Signal Proc, Kuala Lumpur 50603, Malaysia

[2] WeBank AI Lab, Shenzhen, Peoples R China

[3] Hong Kong Univ Sci & Technol, Dept Comp Sci & Engn, Hong Kong, Peoples R China

关键词: Image captioning;    Ownership protection;    Deep neural network;    Recurrent neural network;    Long short-term memory;   

DOI  :  10.1016/j.patcog.2021.108285

来源: Elsevier

【 摘 要 】

By and large, existing Intellectual Property (IP) protection on deep neural networks typically i) focus on image classification task only, and ii) follow a standard digital watermarking framework that was conventionally used to protect the ownership of multimedia and video content. This paper demonstrates that the current digital watermarking framework is insufficient to protect image captioning tasks that are often regarded as one of the frontiers AI problems. As a remedy, this paper studies and proposes two different embedding schemes in the hidden memory state of a recurrent neural network to protect the image captioning model. From empirical points, we prove that a forged key will yield an unusable image captioning model, defeating the purpose of infringement. To the best of our knowledge, this work is the first to propose ownership protection on image captioning task. Also, extensive experiments show that the proposed method does not compromise the original image captioning performance on all common captioning metrics on Flickr30k and MS-COCO datasets, and at the same time it is able to withstand both removal and ambiguity attacks. Code is available at https://github.com/jianhanlim/ipr-imagecaptioning (c) 2021 Elsevier Ltd. All rights reserved.

【 授权许可】

Free