Graduate (List)

This course is designed for students who understand deep learning and convolutional neural networks and studies research topics related to convolutional neural networks. Topics can include residual network, knowledge distillation, network minimization, image-to-image translation, continual learning, domain adaptation, data augmentation, self-supervised learning, meta-learning, attention.

This course provides an overview of neuro-imaging and neuro-image processing techniques. We will discuss the historical and recent development of neuro-imaging technology as well as and neuro-image processing techniques. Neuro-imaging modalities will be discussed with an emphasis on optical imaging. On the imaging processing side, general image processing techniques as well as specialized techniques for processing neuro-images will be discussed.

This lecture provides advanced multimedia processing and learning. Multisensory signal, video, audio, and language are core components of multimedia. Multimodal learning with them is one of the core technologies in multimedia in real-world applications such as intelligent surveillance, smart TVs, and human-machine interface systems. Lecture topics include the basics of multimedia learning, which is image, video, audio, and language representation learning, multimedia fusion schemes, multimedia alignment, and multimedia attention. In addition to the basics of multimodal, students are to participate in term projects and papers reading recently published in multimodal processing and learning.

This course deals with recent techniques based on deep learning for speech pre-processing, recognition, synthesis, and speaker recognition technologies, and makes it possible to extend the area of applications of speech intelligence.

The primary goal of this course is to understand the sound propagation and learn how to extract information from various sound fields. The course covers fundamental principles of sound propagation, beamforming, sound localization, acoustic holography, and related array signal processing techniques for extracting sound information from various environment.

The primary goal of this course is to understand the sound propagation and learn how to extract information from various sound fields. The course covers fundamental principles of sound propagation, beamforming, sound localization, acoustic holography, and related array signal processing techniques for extracting sound information from various environment.

This course introduces selected topics of recent technologies and algorithm related to image processing and imaging systems. (Prerequisite: EE432, EE535)

Image restoration and enhancement problems have been treated as fundamental issues of image processing and computer vision. The image restoration estimates the original (clean) images from their corrupted and noisy image inputs in many forms of motion blur, noise and camera mis-focus etc. Therefore it is performed by reversing the degradation process that causes corrupted images. Different from image restoration, the image enhancement aims at improving the subjective perceptual quality of corrupted images, not necessarily producing the realistic data from a scientific point of view. In general, the image enhancement uses no a priori models of the processes that have created the images.
In recent years, the demand for ultra-high quality images has been increased with the advance in image restoration and enhancement. Especially, deep learning based approaches to image restoration and enhancement have made a great success in performance against the traditional approaches. In this class, we review the traditional approaches to image restoration and enhancement with the analysis of their limitation, and study very recent deep learning based methods using convolution neural networks, recurrent neural networks and generative networks. The students are exposed and experienced to very recent advanced methods in image restoration and enhancement via class lectures and homework assignments and terms projects.

This course is designed to cover the special topics of current interests in electromagnetics.

In the course, basic model architecture and training principles of mostly advanced machine leaning algorithms and their applications will be introduced. The class will cover the Convolution Neural Network, Recurrent Neural Network, Reinforcement Learning as well as transformer network and generative models, especially used in chatGPT. In addition, these advanced transformer models will be used to estimate designs, and to optimize the high-speed and frequency electromagnetic systems, including antenna, circuits, devices, wireless power transfer systems, packages, and semiconductors. These application examples will be followed with discussions.