Agv & Robot Arm Team

1.   

Robot Amm Team

1.   

Robot Arm Team

1.   

Robot Arm Team

1.   

AoI Team

1.   

AoI Team

1.   

FL Team

1.   

FL Team

1.   

Sports Team

1.   

Sports Team

1.   

Bme

1.   

Bme

1.   

Sport

1.   

Bme 醫美

1.   

Bot-Arm

1.   

Bot-Arm

1.   

Bot-Amm

1.   

Bot-Amm

1.   

Sport

1.   

Sport

1.   

Sport

1.   

3D CV Team

1. 基於自注意力神經網路與結構光於單一圖像之3D物體重建   

BME Team

1. 使用改良3D CNN-Transformer模型減少肺結節偵測之偽陽性   

AOI Team & BME Team

1. 使用改良3D Transformer UNet來分割肺部電腦斷層掃描影像之肺結節   

AIot + Sports Team

1. 跨醫院之兩階段肺結節偵測聯邦學習系統   

Robot Amm Team

1. 使用級聯注意力網路於點雲圖中偵測人並結合RTAB-Map   

Robot Arm Team

1. 基於高斯混合模型背景分離或學生老師特徵金字塔圖像匹配應用於鋼筋焊點評估   

Robot Arm Team

1. 基於強化學習使用 CNN-Transformer 模型實現虛實整合之機械手臂夾取   

Robot Arm Team

1. 使用改良DiNAT Transformer和預訓練模型來實現散堆物體抓取系統   

3D CV Team

1. 基於時間卷積網絡及慣性測量單元之電腦斷層導引技術   

BME + Sports Analytics Team

1. 使用改進的Transformer和時空對應網絡之交互式分割技術來標記3D肺結節   

Agv & Robot Arm Team

1.   

Agv & Robot Arm Team

1. 使用三維視覺導引控制機器手臂於自動倉儲系統與使用強化學習於夾取   

3D CV Team

1. 使用改良3D U-Net於肺部血管分割並使用Marching Cube演算法於3D重建，且應用於肋骨   

3D CV Team

1. 使用改良3D U-Net於肺部分割並使用帶注意力模組的孿生神經網路於成對肺圖像匹配   

BME + Sports Analytics Team

1. 在英特爾OpenVINO平台上使用Openpose-LSTM網絡來打造人體骨架的肺部復健系統   

BME + Sports Analytics Team

1. 藉由透視投影和多波段混合方法來實現多相機拼接和自由視角系統   

BME + Sports Analytics Team

1. 使用改良 FairMOT 於游泳比賽之多泳者追蹤、修正和排名   

AOI Team

1.   

AOI Team

1.   

AGV + Robot Arm Team

1. 開發自我監督強化學習技術於免校正抓取系統   

AGV + Robot Arm Team

1. 基於機器人作業系統使用Dual 2D LiDARs建立3D地圖於導航並使用Differentiable RANSAC與CNN於定位   

AGV + Robot Arm Team

1. 使用改良RGB-D MobileNetV3實現平放、曡高和成堆的物體偵測與夾取系統   

BME + Sports Analytics Team

1. 改良YOLOv4模型於英特爾 OpenVINO上來分割肺部電腦斷層掃描影像之亞實質肺結節   

Automatic Optical Inspection

1. 應用深度學習於刀具量測與分類   

Embedded Deep Learning

1. 應用深度學習模板匹配於嵌入式UVW對位平台   

BME + Sports Analytics Team

1. 基於改良YoloV4模型在肺部電腦斷層掃描影像上進行亞實質肺結節檢測   

BME + Sports Analytics Team

1. 使用改良Openpose網路於跳水及游泳之骨架關節估計、修正和分析   

BME + Sports Analytics Team

1. 人工智慧之U-Net架構用於海上油污的光學影像偵測   

BME + Sports Analytics Team

1. Multi-Dots SiamRPN++ Tracking for Pronated Foot Analysis in PSMNet Stereo Space   

BME + Sports Analytics Team

1. Deep Face Recognition using Additive Angular Margin Loss, and Linear Discriminant Analysis in Normalized Feature Weight Space   

AGV + Robot Arm Team

1. 3D Visual-Guided Robot Arm Control for Warehouse Automation System   

AGV + Robot Arm Team

1. Visual-Guided Grasping System using Self-Surpervised Deep Reinforcement Learning   

3D CV Team

1. 3D Panel Alignment System using Deep Learning-Based Structured Light, and System Stability and Speedup   

3D CV Team

1. 3D Object Inspection System Using Deep Learning-Based Structured Light, and System Stability and Speedup   

Automatic Optical Inspection

1.    使用具有微調功能之編解碼器學習網路進行多對焦影像融合

Biomedical Engineering + Sports Analytics

1.    使用具備可分離空洞卷積之編解碼器學習網路進行肺結節分割

Robot Arm Team

1.    用電腦視覺及深度學習來引導機器手臂分辨及抓取螺絲與螺帽

Robot Arm Team

1.    以Faster RCNN 來分離堆疊3D物體的表面區域

Automatic Optical Inspection

1.    在頻譜下運用LSTM於斷刀預期

Automatic Optical Inspection

1.    在頻譜下運用CNN於刀具瑕疵檢測

Embedded Computer Vision

1.    藉由 CNN 及輪廓資料來辨識物體

Embedded Computer Vision

1.    Adaptive Control of Robots withVisual Tracking Controller for Moving Object Interception

Automatic Optical Inspection

1.    在影像下運用Faster RCNN於刀具瑕疵檢測

Deep Learning Team

1.    3D Object Point Estimation by 2D Image Using Multi-Stage Deep Convolutional Neural Networks

Automatic Optical Inspection(鏟花手機APP)

1.    Scraping Inspection Based On HSV Historgram and C.C. Labeling

3D CV Team

1.    Human Joint Estimation and Tracking Using Multi-Stage Convolutional Neural Networks

3D CV Team

1.    3D Point-Cloud Surface Reconstruction by Using KD-Tree-Based Mean Shift Algorithm

3D CV Team

1.    3D Point-Cloud Surface Reconstruction by Using KD-Tree-Based Mean Shift Algorithm

Robot Arm Team

1.    Object Classification and Pick Angle Estimation Using Deep Convolutional Neural Networks for Robot Arm Operation

1.    Reverse Engineering: From 3D Point Clouds to STL to CAD/CAM

1.    Visual-Based Robot Arm Control with Stereo-Based Structured Light

1.    Reverse Engineering: From B-Spline, NURBS to CAD/CAM to 3D Printing

1.    Adaptive Control of Robots with Visual Tracking Controller for Moving Object Interception

2.    Improve Accuracy of Robots with Visual Controller for 2D/3D Calibration

1.    Automatic Optical Inspection (AOI) using Embedded Computer Vision

1.    Reverse Engineering: From B-Spline, NURBS to CAD/CAM to 3D Printing

1.    Tri-Dexel-Based 3D Surface Smoothing and Feature Sharpness Enhancement Using Marching Cube and Workpiece Profile Measurement

1.    Turning Tool Monitoring using Machine Vision

1.    3D Point-Cloud Surface Reconstruction by Using KD-Tree-Based Mean Shift Algorithm

1.    Structured-Light 3D Surface Profilometry Using a Hybrid of Gray Code and Phase Shift Method

1.    Computer Vision for Biomedical Applications

1.    Fabrication Control and Application using DMDMaskless Photolithography Technology

1.    Fast Affine Template Matching using Coarse-to-Fine Optimal Search with Distributed Sampling Points

1.    Self-Organizing Map Neural Network-Based Auto-Focus System for Automatic Optical Inspection

1.    2D-3D Shape and Motion Recovery Using Inverse Compositional Active Appearance Model and Factorization

1.    Circular Hole Center Detection for PCB Alignment by Using Improved Otsu Segmentation and Circle Hough Transform

1.    People Counting Using Time of Flight, Background Subtraction and Hough Detection

1.    Sparse-Based Object Tracking Using Particle Filter

1.    Calibrations for Fisheye Camera and Time-of-Flight (TOF) Sensor

1.    Lane Departure Warning System usingLine Hough Transform and B-spline Curve

1.    Probabilistic Linear Discriminant Analysis-Based Face Recognition using Factor Analysis

1.    SIFT-Based Panoramic Image Stitching with Local and Global Registration

1.    Feature Point Detection

2.    Facial Feature Point Detection using Shape Optimized Search and Tracking using Inverse Compositional Active Appearance Models

1.    Hand Gesture Recognition

2.    Hand Gesture Recognition Based on Dynamic Bayesian Network

1.    Stereo

2.    Maximum-a-Posterior Stereo System Based on Sparse Gradient Point

1.    Face Recognition

2.    Face Recognition Using Tied Factor Analysis

1.    Gender Classification

2.    Local Binary Pattern Histogram-Based Gender Classification Using Real AdaBoost

1.    Multi-View Face Detection

1.    Super-resolution

1.    Crowd Tracking

1.    Fire and Smoke Detection

1.    Face Verification

1.    Gender Classification

1.    People Identification

1.    Road sign Recognition System Based on HOG Descriptors and GentleBoost with Sharing Features

1.    An Adaptive Shifting Vector for Pyramid-Based Video Stabilization

1.    Facial Expression Synthesis: Global Structure and Local Detailed Texture Syntheses Using Factor Analysis and GentleBoost Algorithm

1.    Spatio-temporal Space Learning for Human Action Recognition

1.    Motion Object Detection Using Multiple Parametric Background and Foreground Models at Omni-Directional Camera

1.    License Plate Recognition System for Parking Lots Using Probabilistic Neural Network and 2-layer Ambiguous Character Handling

1.    Global Color Adjustment & Saturation Enhancement for Low-level Image Process

1.    Local and Weighted Maximum Margin Discriminant Analysis

1.    Unsupervised Learning of Discriminative Edge Measures for Vehicle Matching between Non-Overlapping Cameras

1.    Histogram-Based Thin Slide Separation and Count in Solar Cell/Wafer Application

1.    Graphic Model and Intrinsic and Illumination Structures of Nature Images

1.    Color-Based and Gradient-Based Object Tracking Using Particle Filter Embedded in Incremental Discriminant Model

1.    Incremental Perspective Motion Model for Rigid and Non-Rigid Motion Separation

2.    Texture Geometry and Color Regularity Manipulation

1.    COG/COF Particle Detection and Intensity Measurement for TFT-LCD Automatic Optical Inspection

1.    Super-Resolution for Hallucinating Faces

1.    Real-Time Robust Alignment for Small Size COG and COF in Automatic Optical Inspection

1.    Distortion Removal for Digital Photos by Using Boundary Interpolation

1.    Relevance Feedback in Face-Recognition-Based Image Retrieval

2.    Automatic Face Recognition Under Component-Based Manifolds

1.    Real Time Face Tracking by Embedding Mean Shift to Particle Filtering

1.    Fast Tone Reproduction for Digital Images by Using Multi-Resolution Approach

1.    Regularized Bundle-Adjustment to Model Heads from Image Sequences without Calibration Data

1.    Automatic Defect Inspection and Enhancement for Patterned TFT-LCD Panel Surfaces

1.    Automatic Alignment and Inspection of TFT-LCD COG

2.    Efficient Object Segmentation Using Digital Matting for MPEG Video Sequences

1.    Real-Time Facial Feature Detection and Feature Point Location Estimation

2.    Facial Sketch Synthesis Using Direct Combination Model

3.    Facial Occlusion Reconstruction

1.    Image Replacement

2.    Texture Analysis and Texture Synthesis

1.    Animal-like Caricature Generation

2.    Color-Caricature Generation

1.    Rigid and Non-Rigid Motion Segmentation Using Perspective Transformation

1.    Heuristic Pre-Clustering Relevance Feedback in Region-Based Image Retrieval

1.    Semantic Analysis For Tennis Video

1.    3D Reconstruction Using Factorization

1.    Multi-View Face Detection and Pose Estimation

2.    Face Recognition Using Image-Sets

3.    Dimensionality Reduction For Object (Face) Tracking

1.    Graphical Model and Belief Propagation for Machine Vision Applications

2.    Intrinsic and Illumination Structures of Nature Images

3.    Contour Based Detection and Tracking of Moving Objects for Video Surveillance

1.    Face Recognition in Image Sequence

1.    2D Analysis and 3D Synthesis for Facial Expression Analysis

1.    Automatic Vehicle Detection Using Local Features - A Statistical Approach

2.    Real-Time License Plate Detection and Tracking

3.    Facial/Feature Detection and Recognition

4.    Relevance Feedback in Facial-Recognition-Based Image Retrieval

5.    The Sparse Part-Based Object Detection Using The Scale-Invariant Approach

6.    Keypoint (interesting Point) Implementation: Scale Invariant Feature Transform

7.    Keypoint (interesting Point) Implementation: Salient Point

1.   

1.   

1.   

1.   

1.   

1.   

1.   

1.   

1.   

1.   

1.   

1.   

1.   

1.   

1.   

1.   

1.    Augmented Reality Sandbox

1.    Face2Face mapping and animation in real time

1.    Particle Filter Tracking and Panoramic Image Stitching in Analysis of Volleyball Spike

1.    Face2Face mapping and animation in real time

1.    3D LED Display based on Arduino using Kinect for Human-Computer Interaction

1.    Embedded-based Face Recognition for Smart Glassess Application

1.    Embedded-based Face Recognition for Smart Glassess Application

1.    Embedded robot vision using LEGO MindStorms EV3

1.    Embedded robot vision using LEGO MindStorms EV3

1.    Embedded robot vision using LEGO MindStorms EV3

1.    Embedded robot vision using LEGO MindStorms EV3

1.    Remote Robot Control Using Lego MindStorms NXT 2.0

1.    Remote Robot Control Using Lego MindStorms NXT 2.0

1.    Remote Robot Control Using Lego MindStorms NXT 2.0

1.    Remote Robot Control Using Lego MindStorms NXT 2.0

1.    Tennis Game Using Kinect

1.    Tennis Game Using Kinect

1.    Tennis Game Using Kinect

1.    Tennis Game Using Kinect

1.    A Passive Auto-Focus Camera Control System

1.    CAMShift-based Gestural Interface with Colored Finger Cots

1.    CAMShift-based Gestural Interface with Colored Finger Cots

1.    Feature Point Detection

1.    Metric Learning Approaches for Face Identification

1.    A Passive Auto-Focus Camera Control System

1.    CAMShift-based Gestural Interface with Colored Finger Cots

1.    CAMShift-based Gestural Interface with Colored Finger Cots

1.    Feature Point Detection

1.    Metric Learning Approaches for Face Identification

1.    Visibility in Bad Weather from A Single Image

1.    Global Color Adjustment & Saturation Enhancement for Low-level Image Process

1.    Interactive Digital Photomontage

1.    Illumination Normalization with Time–Dependent Intrinsic Images

1.    A High-Resolution Face Sketching by Using AAM and ASM

1.    Incremental Perspective Motion Model for Rigid and Non-Rigid Motion Separation

2.    Texture Geometry and Color Regularity Manipulation

1.    Real–Time Face Detection Using AdaBoost Approach

1.    Automatic Video Tooning

1.    3D Object Reconstruction with Missing Data by Using EM Algorithm

1.    Relevance Feedback in Face-Recognition-Based Image Retrieval

2.    Automatic Face Recognition Under Component-Based Manifolds

1.    Super-Resolution

1.    Shot Detection and Classification in Tennis Video

1.    Real–Time Character Recognition System

1.    Real-Time Facial Feature Detection and Feature Point Location Estimation

2.    Facial Sketch Synthesis Using Direct Combination Model

3.    Facial Occlusion Reconstruction