Presentation | 2021-11-15 [Poster Presentation] A Study on Convergency of DNN Watermarking without Embedding Loss Function Takuro Tanaka, Tatsuya Yasui, Minoru Kuribayashi, Nobuo Funabiki, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Intellectual Property Rights protection related to Deep Neural Networks is an important issue due to the high cost of training DNNs and the widespread use of DNNs in society. We have proposed a method to embed the watermark information directly into the frequency components of some weight parameters without using embedding loss. It has been experimentally shown that the embedding operation has little impact on the convergence of the training. In this paper, we quantitatively evaluate the impact of embedding on the convergence of the training. We also implement the embedding operation on the ResNet in order to check the versatility of the proposed method. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | watermarking / DM-QIM / DNN model / fine-tuning |
Paper # | EA2021-35,EMM2021-62 |
Date of Issue | 2021-11-08 (EA, EMM) |
Conference Information | |
Committee | EMM / EA / ASJ-H |
---|---|
Conference Date | 2021/11/15(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Online |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | [Beginners Session] Engineering/Electro Acoustics, Content Processing, Digital Watermarking, Psychological and Physiological Acoustics, and Related Topics |
Chair | Ryoichi Nishimura(NICT) / Yoshinobu Kajikawa(Kansai Univ.) / 加藤 宏明(NICT) |
Vice Chair | Masaaki Fujiyoshi(Tokyo Metropolitan Univ.) / Masatsugu Ichino(Univ. of Electro-Comm.) / Kenichi Furuya(Oita Univ.) / Shoichi Koyama(Univ. of Tokyo) / 坂本 修一(東北大) |
Secretary | Masaaki Fujiyoshi(Utsunomiya Univ.) / Masatsugu Ichino(NICT) / Kenichi Furuya(NTT) / Shoichi Koyama(RitsumeikanUniv.) / 坂本 修一(東大) |
Assistant | Shoko Imaizumi(Chiba Univ.) / Youichi Takashima(Kaishi Professional Univ.) / Yukou Wakabayashi(Tokyo Metropolitan Univ.) / Tatsuya Komatsu(LINE) |
Paper Information | |
Registration To | Technical Committee on Enriched MultiMedia / Technical Committee on Engineering Acoustics / Auditory Research Meeting |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | [Poster Presentation] A Study on Convergency of DNN Watermarking without Embedding Loss Function |
Sub Title (in English) | |
Keyword(1) | watermarking |
Keyword(2) | DM-QIM |
Keyword(3) | DNN model |
Keyword(4) | fine-tuning |
1st Author's Name | Takuro Tanaka |
1st Author's Affiliation | Okayama University(Okayama Univ.) |
2nd Author's Name | Tatsuya Yasui |
2nd Author's Affiliation | Okayama University(Okayama Univ.) |
3rd Author's Name | Minoru Kuribayashi |
3rd Author's Affiliation | Okayama University(Okayama Univ.) |
4th Author's Name | Nobuo Funabiki |
4th Author's Affiliation | Okayama University(Okayama Univ.) |
Date | 2021-11-15 |
Paper # | EA2021-35,EMM2021-62 |
Volume (vol) | vol.121 |
Number (no) | EA-246,EMM-247 |
Page | pp.pp.49-54(EA), pp.49-54(EMM), |
#Pages | 6 |
Date of Issue | 2021-11-08 (EA, EMM) |