國科會研究計畫
NSTC Research Projects

1. 基於量 化、剪枝與架構搜索之 AI 模型壓縮技術研究
-
合作對象:國科會(學術研究)
-
技術領域:AI 模型優化
-
內容:研究如何透過量化、剪枝和神經網絡架構搜索等方法來縮小 AI 模型的體積與複雜度。
-
AI Model Compression via Quantization, Pruning, and Architecture Search
-
Partner: NSTC (Academic Research)
-
Technology: AI Model Optimization
-
Description: Researched methods to reduce AI model size and complexity through quantization, pruning, and neural architecture search.
-

2. 應用基於深度學習技術之影像指紋技術偵測影片竄改
-
合作對象:國科會( 學術研究)
-
技術領域:電腦視覺、深度學習
-
內容:開發基於深度學習的影像指紋技術,用以辨識影片是否經過不當修改或偽造。
2. Deep Learning-Based Image Fingerprinting for Video Tampering Detection
-
Partner: NSTC (Academic Research)
-
Technology: Computer Vision, Deep Learning
-
Description: Developed image fingerprinting technology to detect unauthorized video modifications.

3. 應用基於深度增強式學習之自我組織網路在網路切片、異質網路及裝置對裝置問題上研究
-
合作對象:國科會(學術研究)
-
技術領域:深度增強式學習、通訊網路
-
內容:將深度增強式學習應用於自我組織網路,以解決網路切片、異質網路整合及 D2D 通訊等挑戰。
3. Deep Reinforcement Learning for Self-Organizing Networks
-
Partner: NSTC (Academic Research)
-
Technology: Deep Reinforcement Learning, Communications Networks
-
Description: Applied deep reinforcement learning to self-organizing networks for network slicing, heterogeneous networks, and device-to-device communication challenges.

4. 以大數據技術為基礎之次世代行動通訊自我組織網路機制之研究
-
合作對象:國科會(學術研究)
-
技術領域:大數據、通訊網路
-
內容:研究如何利用大數據技術,為下一代行動通訊網路建立自我組織與優化的機制。
4. Big Data-Based Self-Organizing Mechanisms for Next-Generation Mobile Networks
-
Partner: NSTC (Academic Research)
-
Technology: Big Data, Communications Networks
-
Description: Researched big data techniques to build self-organizing and optimized mechanisms for next-generation mobile networks.

5. 設計與實作基於 Transformer 之語音轉語音模型協助失聾人士實現語音轉換
-
合作對象:國科會(學術研究)
-
技術領域:語音處理、深度學習
-
內容:開發一個基於 Transformer 架構的語音對語音轉換模型,旨在幫助聽障人士進行語音溝通。
5. Transformer-Based Speech-to-Speech Model for Deaf Individuals
-
Partner: NSTC (Academic Research)
-
Technology: Speech Processing, Deep Learning
-
Description: Developed a Transformer-based speech-to-speech model to assist deaf individuals with communication.

6. 基於 Transformer 模型之圖像轉語音視障人士輔助系統
-
合作對象:國科會(學術研究)
-
技術領域:電腦視覺、語音處理、深度學習
-
內容:開發一個能將圖像內容轉換為語音描述的輔助系統,以協助視障人士理解周遭環境。
6. Transformer-Based Image-to-Speech System for the Visually Impaired
-
Partner: NSTC (Academic Research)
-
Technology: Computer Vision, Speech Processing, Deep Learning
-
Description: Created a system to convert image content into speech descriptions to aid visually impaired individuals.
