MMM 2020 26th International Conference on Multimedia Modeling January 5-8, 2020 / Daejeon, Korea

Tutorials

Jan 5(Sun.) 2020 Tutorial Ⅰ
13:00-14:30
Session title Introduction to biometrics and anti-spoofing
Organizer Wonjun Kim (Konkuk university)

There has been a notable progress in biometrics with the improvement of sensors in the last few years, and some traits e.g., face, fingerprint, and iris, now start to be applied to mobile devices for payment as well as security. Even though the high-performed approaches only using images have achieved a mature level for the recognition performance (e.g., face recognition : >99% on LFW dB), most of them still suffer from various spoofing attacks. For example, printed photos, mimic masks, and screen shot of a valid user can be easily obtained from public domains (for example, internet) and those are readily employed for malicious log-in attempts. in this talk, representative methods for face, fingerprint, iris, and vein recognition proposed in literature are briefly introduced. Moreover, state-of-the-arts for anti-spoofing including deep neural network-based approaches, are also explained with the performance comparisons based on various benchmark dBs.

Tutorial Ⅱ
14:30-16:00
Session title Recent advances in deep novelty detection for medical imaging
Organizer Jaeil Kim (Kyungpook national university)

Novelty detection is the task of identifying whether test data is an outlier from the training data in some aspects. approaches for the novelty detection exploit explicit representation of the distribution of the training data (i.e., positive samples) to determine outliers in feature space. unseen samples are compared with the models of normality, which can draw the decision boundary of the training samples. in medical imaging, the novelty detection has gained much attention, due to the lack of sufficient dataset for diseases and the limitations in annotating medical images for supervised learning. Recently, deep auto-encoder methods have been proposed to localize abnormal regions, such as multiple sclerosis and brain tumor, in brain MR images. Weakly-supervised anomaly detection methods using epistemic uncertainty based on bayesian deep networks also showed better performance in characterizing anomalies under several disease conditions in medical images. in this tutorial, we will review recent advances in the novelty detection by deep learning methods and introduce their challenges in medical imaging.

Tutorial Ⅲ
16:00-18:00
Session title Haptic interaction with multimedia data
Organizer Sang-Youn Kim (Korea university of technology and Education)

Haptic display and rendering has emerged as a core application area in multimedia. in this tutorial, we present fundamental knowledge for a haptic display, haptic rendering, and associated evaluation methods. in particular, this lecture focuses on the important concepts in haptic perception and hCI. Real examples where such techniques are applied are also introduced to help understanding.