IEICE Technical Report

Online edition: ISSN 2432-6380

Enriched MultiMedia

Workshop Date : 2026-03-05 - 2026-03-06 / Issue Date : 2026-02-26

EMM2025-116
Development and Evaluation of a Voice Training System to Promote Evacuation -- Feedback from Deep Learning-based Scoring --
Sayoko Takano (Kanazawa Inst. Tech.), Zen Nagatsuka (Zen Voice Factory), Yosio Tsuchida (Kanazawa Inst. Tech.)
pp. 1 - 5

EMM2025-117
Investigation of discrimination of pitch and tone-height for vowels and words
Shota Morita (Fukuyama Univ.), Teruki Toya (Univ. Yamanashi)
pp. 6 - 10

EMM2025-118
Estimation of autonomic nervous system activity while viewing Noh performances using skin potential measurements
Shunsuke Kidani (JAIST)
pp. 11 - 14

EMM2025-119
[Poster Presentation] Detecting Real or Fake Videos with Wavelets and Explainable AI
Ryouta Komatsu, Hyunho Kang (NITTC)
pp. 15 - 18

EMM2025-120
[Poster Presentation] Matrix-Multiplication-Based White-Box DNN Data Hiding Robust Against Desynchronization Attack
Shinichiro Tokuda, Shoko Imaizumi (Chiba Univ.), Minoru Kuribayashi (Tohoku Univ.)
pp. 19 - 24

EMM2025-121
[Poster Presentation] A Dual Obfuscation Method for Federated Learning to Enhance Robustness against Image Restoration Attacks
Yuki Itabashi, Hiroto Sawada, Mare Hirose, Shoko Imaizumi (Chiba Univ.), Hitoshi Kiya (Tokyo Metropolitan Univ.)
pp. 25 - 30

EMM2025-122
[Poster Presentation] Evaluation of a Watermarking Technique for Vision Transformer
Junya Yamamoto, Shigeyuki Sakazawa (OIT)
pp. 31 - 35

EMM2025-123
[Poster Presentation] Improved Loss Function for Watermark Embedding in VQ-VAE
Masato Takimoto, Shigeyuki Sakazawa (OIT)
pp. 36 - 40

EMM2025-124
[Poster Presentation] Single-Layered Variable-Length Encoding Networks for DiffuseTrace and Its Performance Evaluation
Ou Egami, Masaki Kawamura (Yamaguchi Univ.)
pp. 41 - 46

EMM2025-125
[Poster Presentation] Defense against Model Extraction Attacks on Diffusion Models by Embedding Inheritable Watermarks into Generated Images
Hayato Shoji, Kazuaki Nakamura (Tokyo Univ. of Science)
pp. 47 - 52

EMM2025-126
[Poster Presentation] A Proposal for Trigger Set Generation Method for Blackbox Watermarking under Luminance Constraints
Taichi Morinaka, Masaki Kawamura (Yamaguchi Univ.)
pp. 53 - 58

EMM2025-127
[Poster Presentation] A Study on Deep Learning-Based Person Tracking in Surveillance Videos
Hiroki Sano, Hyunho Kang (NITTC)
pp. 59 - 64

EMM2025-128
[Poster Presentation] AI Model for Predicting Machine Part Failures
Tomoki Makita, Hyunho Kang (NITTC)
pp. 65 - 69

EMM2025-129
[Poster Presentation] Hand Gesture Recognition as a Learning Tool: Using MediaPipe for Teaching Binary, Hexadecimal, and Digital Logic Concepts
Papu Saroma, Hyunho Kang (NITTC)
pp. 70 - 75

EMM2025-130
[Poster Presentation] Study on Emotion Estimation in Conversational Speech Data Using Sound Quality Metrics
Natsuki Nagao, Shogo Okada, Masashi Unoki (JAIST)
pp. 76 - 81

EMM2025-131
[Poster Presentation] Technology for Protecting the Authenticity of Voice AI Content and User Experience Design for Social Implementation
Shimose Koichi, Nana Nishimura, Daisuke Horibe (NTTWest)
pp. 82 - 85

EMM2025-132
[Poster Presentation] Subjective evaluation of stereo width shrinkage method using SD method
Arisa Matsumoto, Mitsunori Mizumach (KIT)
pp. 86 - 90

EMM2025-133
[Poster Presentation] Classification of AI-generated Images using Reconstruction Error on Sparse Coding
Shoto Yamashita, Michiharu Niimi (KIT)
pp. 91 - 96

EMM2025-134
[Poster Presentation] Studies on protecting Style Mimicry by Text-to-Image Models based on simple noise addition in latent space
Shinki Furukawa, Michiharu Niiimi (KIT)
pp. 97 - 102

EMM2025-135
[Poster Presentation] Deep Learning-based Fall Detection from Surveillance Videos
Taiki Abe, Hyunho Kang (NITTC)
pp. 103 - 107

EMM2025-136
LYRIC-DRIVEN SLIDESHOW GENERATION SYSTEM VIA CROSSMODAL RETRIEVAL USING TEXT-COLOR MULTIMODAL FEATURES
Mei Hashimoto, Michiharu Niimi (KIT)
pp. 108 - 113

EMM2025-137
Blind estimation of room acoustic parameters from observed speech signals.
Masashi Unoki (JAIST)
pp. 114 - 119

EMM2025-138
A Study on Vulnerability to Generative AI-Enabled Fraud
Shuji Tonomura, Kazuhiro Kono (Kansai Univ.)
pp. 120 - 125

EMM2025-139
(See Japanese page.)
pp. 126 - 127

Note: Each article is a technical report without peer review, and its polished version will be published elsewhere.

The Institute of Electronics, Information and Communication Engineers (IEICE), Japan