Smart Multimedia 2025, Program
***********************************
Day 1 - Friday, Dec. 19, 2025
***********************************
8:45 – 9:00 Opening Ceremony (Sylvain Bouchigny, Arnuad Leleve and Carlos Rossa)
======================================================================================================================================
9:00 – 10:20 Oral Session 1 (OS1): Machine Learning for Multimedia / Haptic – Chair Christian Bolzmacher:
======================================================================================================================================
9:00 – 9:20 Development of a Rotary Knob with Haptic Feedback Employing Ultrasonic Lubrication Via In-Plane Modes of a Piezoelectric Ceramic Actuator
Christian Bolzmacher, Gwénaël Changeon, Laurent Eck (France)
9:20 – 9:40 HistDiff: Diffusion Models Empowered by Histogram Matching for Reducing Content Color Interference in Style Transfer
Siying Liu, Zhenye Liu, Qiaoanna Konglang, Yingnan Ma, Anup Basu (Canada)
9:40 – 10:00 Performance Analysis of Deep Learning Architectures for Furnishing Status Classification
Oguzhan Guclu, Muhammet Güven, Irem Islek, Çağla Çığ Karaman (Turkey)
10:20 – 10:40 Multi-Temporal Latent Diffusion Transformers for Cloud Removal in Remote Sensing Images
Faisal Ahmed, Cameron Jen, Anup Basu (Canada)
10:20 – 10:50 Coffee Break
======================================================================================================================================
10:50 – 12:10 Oral Session 2 (OS2): Medical Applications – Chair Lynda Bounif:
======================================================================================================================================
10:50 – 11:10 GAN-Boosted U-Net for Enhanced Breast Tumor Segmentation: A Hybrid Approach for Improved Performance and Interpretability
Shokofeh Anari, Ramin Ranjbarzadeh, Malika Bendechache (Ireland)
11:10 – 11:30 SWARMS - Smart Wrist-Based Assembly Recognition and Monitoring System for Small Parts (special session)
Salsabil Soliman, Vishnu Kakaraparthi, Troy McDaniel (USA)
11:30 – 11:50 Comparative Robustness Analysis of ResNet (V1, V2, RS) and BoTNet Models for ALL Diagnosis from Compressed Peripheral Blood Smears
Lynda Bounif, Amina Serir (Algeria)
11:50 – 12:10 Passive Haptic Learning for Different Information Types
Connor Giam, Troy McDaniel (USA)
12:00 – 14:00 Lunch
======================================================================================================================================
14:00 – 15:20 Oral Session 3 (OS3): Multimedia for Humans – Chair Iulia Cristina Stanica:
======================================================================================================================================
14:00 - 14:20 A Multisensory Approach to Rehabilitation: Smart Multimedia and Gamification in Virtual Reality
Iulia Cristina Stanica, Giovanni-Paul Portelli (Romania)
14:20 – 14:40 A Deep Learning-Based Multi-Feature Fall Detection Approach Using Silhouette and Motion Data
Qilong Yu, Letian Xu, Siming Hua (Canada)
14:40 – 15:00 Speech As a Multimodal Digital Phenotype for Multi-Task LLM-Based Mental Health Prediction
Mai Ali, Christopher Lucasius, Tanmay Pranav Patel, Madison Aitken, Jacob Vorstman, Peter Szatmari, Marco Battaglia, Deepa Kundur (Canada)
15:00 - 15:20 Energy-Based Body Reconstruction from Sparse Image Sequences Using Single Depth Camera
Jeffrey Qiu, Deepa Kundur (Canada)
15:20 – 15:50 Coffee Break
======================================================================================================================================
15:50 – 16:50 Oral Session 4 (OS4): Haptic – Chair Sathiyamoorthy Selladurai:
======================================================================================================================================
15:50 – 16:10 Uncertainty-Aware Post-Detection Framework for Enhanced Fire and Smoke Detection in Compact Deep Learning Models
Aniruddha Srinivas Joshi, Shreyas Srinivas Joshi, Godwyn James William (USA)
16:10 – 16:30 Artifact-Resilient Needle Visualisation in Prostate Brachytherapy Using Angular Plane Wave Compounding and Confidence-Weighted Imaging
Sathiyamoorthy Selladurai, Hoorieh Mazdarani, Carlos Rossa (Canada)
16:30 - 16:50 Consideration of the Effect of Different Weight of the Object on the Change in Force Perception When Applying Vibration to the Wrist During Assisted Lifting Motion (Japan)
Tatsuya Kiyoshi, Kazunori Ogawa, Yuichi Kurita (Japan)
***********************************
Day 2 - Saturday, Dec. 20, 2025
***********************************
======================================================================================================================================
9:00 – 10:20 Oral Session 5 (OS5): Image / Video – Chair Jiying Zhao:
======================================================================================================================================
9:00 – 9:20 Temporal Pyramid Structure for Video Frame Interpolation
Jiaqi Yang, Yu Liu, Jiying Zhao (Canada)
9:20 – 9:40 RAST 4.0: Restorable Arbitrary Style Transfer via Content Leakage Correction
Yingnan Ma, Xudong Li, Anup Basu (Canada)
9:40 – 10:00 Segmentation of Cars Covered with Smoke
Dageng Ding, Yu Liu, Jiying Zhao (Canada)
10:00 – 10:20 Toward Real-Time 4D Animation for Multiple Surveillance Cameras with AI Vision
Chao-Lieh Chen, Yi Hong Zhan, Chang-Hui Chang (Taiwan)
10:20 – 10:50 Coffee Break
======================================================================================================================================
10:50 – 12:10 Oral Session 6 (OS6): Medical Applications 2 – Chair Martyna Fröhlich:
======================================================================================================================================
10:50 – 11:10 Synergistic Fusion of Multimodal Data for Automated and Accurate Diagnosis of Parkinson's Disease
Daria Hemmerling, Martyna Fröhlich, Marta Kaczmarska, Miłosz Dudek, Justyna Krzywdziak, Kamil Burza, Marek Wodziński (Poland)
11:10 – 11:30 SWARMS - Smart Wrist-Based Assembly Recognition and Monitoring System for Small Parts (special session)
Salsabil Soliman, Vishnu Kakaraparthi, Troy McDaniel (USA)
11:30 – 11:50 Deep Learning-Based Image Fusion with Novel Multi-Focus Dataset for Extended Depth of Focus in Microscopy
Ramazan Ozgur Dogan, Fatma Tuana Dogu, Noemia Santos de Oliveira, Imane Kannada, Hulya Dogan, Sena Sezan (Turkey)
11:50 – 12:10 Spatial Audio-Based Exercise Support System - Evaluation of Sound Source Localization Accuracy
Issei Kikuchi, Ariaki Higashi, Naoko Shinagawa, Yuichi Kurita (Japan)
12:00 – 14:00 Lunch
======================================================================================================================================
14:00 – 15:20 Oral Session 7 (OS7): Vision – Chair Maximilian Wachter:
======================================================================================================================================
15:00 – 15:20 Transformer-Based Rhythm Quantization of Performance MIDI Using Beat Annotations
Maximilian Wachter, Sebastian Murgul, Michael Heizmann (Germany)
14:00 – 14:20 A Lightweight Deep Learning and Data Augmentation Framework for Cattle Weight Estimation under Unstable Camera Conditions and Small Datasets
Sachin Vijay Kumar, Xinyao Sun, Irene Cheng (Canada)
14:20 – 14:40 Modeling Heteroscedastic Noise in Infrared Thermography Using Expectation-Maximization PCA
Jasleen Kaur, Vanita Arora, Prabhu Babu, Ravibabu Mulaveesala (India)
14:40 – 15:00 ReSimX: A Local Feature-Aware Framework for Real2Sim and Sim2Real Transfer
Siying Liu, Yingnan Ma, Zhenye Liu, Qiaoanna Konglang, Chenqiu Zhao, Anup Basu (Canada)
15:20 – 15:50 Coffee Break
======================================================================================================================================
15:50 – 16:50 Oral Session 8 (OS8): Temporal analysis – Chair Zohre Foroushi:
======================================================================================================================================
15:50 – 16:10 MetaErr: Towards Predicting Error Patterns in Deep Neural Networks
Varun Totakura, Shayok Chakraborty (USA)
16:10 – 16:30 TITR-CNN: Temporal Inpainting with Trend–Remainder Decomposition for Partially Occluded Lip-Region Video Reconstruction
Zohre Foroushi, Richard Dansereau (Canada)
16:30 – 16:50 Comparing Pose Estimation Models for Lower Body Joints
Bhavya Sehgal*, Thomas Truong, Irene Cheng (Canada)
19:30 – 22:00 Banquet
***********************************
Day 3 -Sunday, Dec. 21, 2025
***********************************
Sightseeing