A DL engineer with extensive experience in computer vision and NLP projects. Strong in image signal processing, building 3D simulation and data analysis. Highly creative in development and proficient in improving deep learning models with novel ideas.Machine Learning/Deep LearningComputer vision: image siganl processing, classification, object detection, semantic segmentation, domain generalization, transfer learningNatural language Processing: recommendation system, information retrieval, question answeringData PreprocessingImage preprocessing: image augmentation, satellite image preprocessing, data crawling (ArcGIS)Text preprocessing: data crawling, tokenization, text cleaningSoftware DevelopmentAPI development, SDK development, Acrobat PDF-based platform development3D games, virtual reality application, 3D animation, real-time 3D visualization and simulation (Unity)Database ManagementObject oriented database: information management of 3D modeling (Revit, MicroStation)Relational database: Acrobat PDF embedded database managementStructural Design & AnalysisNumerical analysis, structural analysis, seismic design and analysis (ETABS)Reinforcement concrete structure design, steel structure design, structural reinforcementProgrammingPython, Pytorch, JavaScript, C++, HTML, MATLAB, C#LanguageMandarin (native), English (excellent). Korean (fluent)
-
Camera System EngineerQualcomm Apr 2022 - Apr 2024Taipei, Taipei City, TaiwanISP module tuning- Tune modules for better texture and noise performances especially in low light scenes.- Tune modules for better chroma and luma performances.Feature validation and performance evaluation- Refine feature calibration process.- Develop auto IQ evaluation tool for efficient feature test. -
Dl Research Fellow (Ai Research Center, Computer Vision Project)National Taiwan University Jan 2021 - Apr 2022Taipei City, TaiwanApplication of Remote Sensing Image and Deep Learning Technique for Disaster Management- Preprocessed satellite images by ArcGIS and the GDAL library to split images and generate raster labels.- Adopted the attention mechanism to catch non-local features of road network.- Improved generalization of the proposed model with single domain training data.- Assisted post-disaster response with extracted accessible road network. -
Research Assistant (Computational Finance And Data Analytics Lab, Nlp Project)Academia Sinica, Taiwan Mar 2020 - Dec 2020Taipei City, TaiwanConversational Assistance Model in Information Searching- Identified the ellipsis problem where words are omitted in given conversational queries.- Translated input sentences into possible original questions by converting latent queries to expand queries.- Solved the ellipsis problem by expanding queries with latent queries derived from related passages.Cross domain recommendation system for the cold-start video-on-demand (VOD) users- Analyzing users’ preferences with design of experiment (DOE).- Improved VOD recommendation system based on users’ short-term and long-term VOD preferences.- Alleviated the cold-start problem by utilizing the descriptions of TV programs on users’ records. -
Bim Engineer & Platform DeveloperIdm Co.: Infrastructure Design & Management Jan 2018 - Jan 2020South KoreaPlatform Development for Information Management and Visualization- Designed user interface which integrated structural analysis and 3D modeling in a collaborative real-time platform.- Developed a PDF-based platform for cable bridge construction management
-
Graduate Research Assistant (Construction And Ml Lab)Seoul National University Sep 2016 - Jun 2017Seoul, South KoreaAutomatically Building As-Built BIM from Point Cloud Model- Researched the point cloud model that can accurately and automatically model the existing structure layout.- Utilized data mining method to extract the interesting structural elements. -
Master ThesisNational Taiwan University Jul 2015 - Jul 2016Taipei City, TaiwanEvaluation of Shear Strength and Design Procedure for Steel Fiber Reinforced Concrete Exterior Beam-Column Joints- Proposed a shear strength model to calculate the shear strength of steel fiber reinforced concrete (SFRC) beam-column joints with the average 15% error.- Investigated at least 30% of transverse reinforcement can be reduced (compared with the amount ACI acquired) while applying sufficient steel fibers by using MATLAB. Expected to ease the congestion problem with the great reduction of transverse reinforcement.- Proposed a detailed design suggestion for SFRC joints based on the proposed shear model and quantified efficiency.
-
Global Project For Temporary Structures In Post-Disaster RegionsNational Taiwan University Jan 2015 - Jun 2015Taipei City, Taiwan- Led a global team, constituted by 15 students from the USA, India, Netherlands, South Korea and Taiwan, to design a construction project for quickly rebuilding temporary and permanent structures after a flood disaster.- Successfully managed the project progress and appropriately arranged the sub-works to members, even though the members joined the project in the different period.- Well held the meetings and discussions. With full communications, great cooperation was awarded.
-
Project Of Visualization Of The Autonomous Beam Assembly SystemNational Taiwan University Feb 2014 - Jun 2014Taipei City, Taiwan- The visualization simulates an autonomous beam assembly system (ABAS) which is developed to prevent the workers from working in the high place.- The rotation method is a flywheel equipped at the top of the rigging beam aiming to rotate the beam to the assembly angle.- The alignment method includes both vertical and horizontal alignments. The vertical alignment relies on a camera, a marker on the column, and a light signal to align the beam altitude. The horizontal alignment relies on the special-designed shape that can smoothly guide the beam into the right position.- The bolting method is used to assembly the beam. The bolt can plug in and slide to the bolt hole.- The unloading method is used to unload the cable and the ABAS
Yu-Hsuan Tseng Education Details
-
Civil Engineering Structural Engineering Program
-
Exchange Student
-
Civil Engineering
Frequently Asked Questions about Yu-Hsuan Tseng
What is Yu-Hsuan Tseng's role at the current company?
Yu-Hsuan Tseng's current role is Diverse deep learning model development in computer vision and NLP products. Commit to developing applications integrating deep learning techniques with multimodal data..
What schools did Yu-Hsuan Tseng attend?
Yu-Hsuan Tseng attended National Taiwan University, Seoul National University, National Taiwan University.
Not the Yu-Hsuan Tseng you were looking for?
-
-
-
-
2inventec.com, gmail.com
Free Chrome Extension
Find emails, phones & company data instantly
Aero Online
Your AI prospecting assistant
Select data to include:
0 records × $0.02 per record
Download 750 million emails and 100 million phone numbers
Access emails and phone numbers of over 750 million business users. Instantly download verified profiles using 20+ filters, including location, job title, company, function, and industry.
Start your free trial