Intelligent Multimedia Applications

Lip-Reading Recognition (Landseed Hospital Collaboration)

Visual lip-reading recognition for assistive and clinical use, developed with Landseed International Hospital (聯新國際醫院).

Multispectral UAV Imagery AI for Smart Agriculture

A MobileNetV2-based multispectral farmland-classification processor for unmanned-aerial-vehicle platforms, supporting smart-agriculture remote sensing.

G2LGAN:對不平衡資料集進行資料擴增應用於晶圓圖缺陷分類

Semiconductor manufacturing requires multiple complex chemical processes. Errors in any link will cause defects in the produced wafers. Therefore, wafer map defect classification is a key task for the semiconductor industry to maintain and improve yield.....

SeeDetail　

應用於遠距醫療之無損心電壓縮系統

隨著社會高年齡層的老化，慢性病的人口逐漸增加，因此居家看護系統即成為未來醫療保健中重要的一環。針對心血 管慢性疾病人口的成長和個人居家醫療照護的需求，以及網路雲端的興起，遠距醫療照護將 成為未來發展趨勢。

SeeDetail 　

口罩人臉辨識與體溫測量系統

隨著COVID-19的爆發，防疫是防止流行病蔓延的一種方式。 COVID-19 的典型症狀包括發燒、咳嗽和呼吸困難。因此，醫院、學校、辦公場所等很多公共場所都需要消毒和測溫。...

SeeDetail　

跌倒偵測系統

SeeDetail　

大人小孩

SeeDetail　

應用於智慧廚房家電之孩童警報系統

物聯網（IoT）是一種將所有物品串連在一起的智能網絡，加入物聯網的智能科技可 以使日常物品成為“可思考和溝通的裝備”。 以此為構想起點，在電視新聞常看到如嬰兒在廚房打翻水瓶被熱水燒燙傷或小孩在廚房玩鬧家電造成大火災等家庭 廚房意外事件。 為了杜絕居家安全事件再度發生，我們開發一 套配合攝影機和行動裝置的廚房安全監控系統，此套系統透過物聯網技術使廚房中 危險家電賦予智慧的功能，快速辨認小孩是否靠進廚房裡的危險家電，讓不在廚房的大人可以立刻得知，以保護我們的小孩。

SeeDetail 　

Variable-Rate Transformer-CNN Hybrid Adaptive Image Compression

A variable compression-rate framework that fuses Transformer and CNN features for adaptive image compression, balancing bit-rate and reconstruction quality within a single model. (M.S. thesis, 2025)

Structure-Guided Image Restoration with NSVQ-VQVAE and KSVD-like Codebook

A structure-guided image restoration scheme combining an NSVQ-VQVAE with a KSVD-like codebook recovery mechanism to reconstruct fine image details. (M.S. thesis, 2025)

StripWinformer: Locally-Enhanced Transformer for Image Motion Deblurring

Single image motion deblurring is a critical low-level computer vision task, aiming to restore clear and sharp images from motion-corrupted counterparts. Traditional image motion deblurring methods often face challenges in handling complex motion patterns and preserving fine details. Inspired by the success of vision transformer models in various tasks, we propose an innovative framework.....

SeeDetail