research interests by wen-hung liao 廖文宏 9/27/2007
Post on 15-Jan-2016
234 views
TRANSCRIPT
Research Interests
By Wen-Hung Liao
廖文宏9/27/2007
Focus of VIPL Research
Multimedia signal processing, including: Video processing Image analysis Audio signal processing
A common thread: human perception of the physical world. (cognitive multimedia processing)
Put more emphasis on content analysis than compression/multimedia format/standards.
Video Processing Semantic analysis of sports video:
swimming style classification Real-time video effects:
compressed-domain Non-Photorealistic rendering (NPR)
Face detection and tracking: (see also DDPlayCam) Facial expression analysis using local appearance i
nformation Video-based human computer interaction: Magic Mir
ror Video analysis in sleep studies Motion tracking and analysis using real-time stereo Skin color segmentation using achromatic features
Swimming Style Classification
Objective: to classify swimming motion into four styles, namely, Backstroke (仰式 ) Breaststroke (蛙式 ) Butterfly (蝶式 ) Free style (自由式 )
using video recorded with an above-water camera.
Demo
Compressed-Domain NPR
Original video With oil-paint effect
Face Detection and Tracking
Facial Expression Analysis
Image Analysis Biometrics:
Face recognition under low illumination Recognition of human faces with glasses
Automatic caricature generation (2D and 3D) Textured-image-based CAPTCHA (Completely
Automated Public Turing Test to Tell Computers and Humans Apart: Examples)
Image-based CAPTCHA Attention-based personal photo manager Eye tracking Recognition of Hand-drawn Geometric Objects
2D Caricature
Basic Idea
畫家完稿 輸入相片 自動產生之肖像畫
=+
2D Caricature: Some Results
+
=
3D Caricature
3D Caricature: Illustration
Textured-image-based CAPTCHA
Static Patterns:
Dynamic
Image-based CAPTCHA
Easy
Hard
Attention-Based Photo Manager
影像資訊 影像瀏覽器 影像重要性
EXIF 資訊 影像分類
提供使用者參考 提供索引與搜尋
資訊融合
1.個人紀念照
2.個人大頭照
3.團體紀念照
4.純粹風景照 對焦品質 曝光品質 使用者專注
影像品質評估 使用者行為評估
瀏覽時間
Gaze Trajectories
Gaze Path Comparison
Audio Signal Processing
Voice User Interface based on VoiceXML Personalized Information Broadcasting Syste
m Birdcall recognition News audio evaluation and analysis Snoring analysis
Comments/Questions?