relationship maintenance 24 Merchants, 6 IPs. NFT generation and 3D modeling Web 3.0, Metaverse, NFT-related consultant. OctFeb 2021 Project Engineer Acer Studying cutting-edge AI/DL skills to implement on a medical AI project. 92% accuracy on glaucoma CDR detection. Familiar with object detection, segmentation, and classification AI scenario. Good communication skills with doctors' demands and collaboration with colleagues. Patent Disclosure: Ultrasound detect and notify system. (serial number: I學歷 SepJun 2 National Taiwan University of Science and Technology Masters in Electrical Engineering Thesis "Online Data Stream Analytics

Research Experience Research Assistant, Seppresent Communication and Multimedia Lab (CMLab), NTU Proposed a Convolutional Neural Network (CNN) accurately performing fine-grained classification for surveillance car. The model recovered lost details from low resolution image with hints in crawled web images. Leveraged frame similarity to speed up CNN-based object detection and semantic segmentation models, e.g. FPN, PSPNet, YOLO, Faster-RCNN, and SegNet. Proposed a multimodal CNN achieving high fine-grained classification accuracy. The model was trained with both web images and its tags, and could predict solely with image in future testing phase

限公司 Java/ Kotlin/ python/ Flutter/ Docker/ Django/ FastAPI/ MySQL/ AWS/ GCP/ OpenCV - Direct management of about 5 to 8. - Mobile app development about AR, temperature/co2 monitoring, beacon parking-related app, etc. - Computer vision for object detection, hand tracking, hair segmentation, and virtual makeup with python language using OpenCV and other open-source. - The assessment of new technology to import and manage engineers' working flow. - Web API development for mobile or Web, and deploy with Docker container on a cloud server or local

用 6 種不同 CNN model 搭配 5 種 data augmentation 並且 fine-tune model，達到 93 % Accuracy on validation dataset. Accuracy 達到 93%, 已與醫師合作發表醫學 paper 3. Pathology案件 -建立 two stage segmentation model, stage one segmentation model 達到86% IOU, stage two segmentation model 達到 94% custom dice coefficient 學歷 清華大學 動力機械工程學系技能 Software AWS Airflow Dagster Docker Git Flask Pytorch Tensorflow DVC Languages Python SQL Bash

Detection: Detect Player Contacts from Sensor and Video Data, 2023/03/03 AIGC (生成式AI): - Bronze medal (solo): (Kaggle) Stable Diffusion - Image to Prompts: Deduce the prompts that generated our "highly detailed, sharp focus, illustration, 3d renders of majestic, epic" images, 2023/05/16 Object detection (目標檢測): - 4th place (solo): (Aidea AI CUP) 肺腺癌病理切片影像之腫瘤氣道擴散偵測競賽 I：運用物體偵測作法於找尋STAS, 2022/06/02

ensuring the timely recording and distribution of clients' rewards. •Drafted, proofread, designed, and distributed both internal and external communications. This included visually compelling Google Display Ads, press releases, engaging blog posts, and posts across multiple social media platforms. •Maximized the use of the CRM system for customer segmentation and targeted marketing. •Coordinated promotional items for prospects and customers, leveraging digital platforms to achieve a wider reach and impact. •Engaged with internal teams to optimize procedures and campaigns that fueled growth marketing initiatives. •Collaborated closely with product and engineering teams to implement strategies aimed at

s project has been able to grab a variety of daily necessities, including fried chicken legs and bundled wires (flexible objects). Detecting the defects of industrial products Tensorflow、Segmentation、Unet Cooperate with well-known factories and manufacturers to detect the defects of various items. Traditional algorithms need to design multiple layers of logic and processing for complex images, but general-purpose neural networks can overcome this problem. Design a general-purpose model, so that the model (Segmentation) can be generalized and learned effectively. The objects include 17 kinds of items, such as ti...