Research
- we have open positions for PhD students in the area of Image and Object Recognition and/or Sign Language Recognition and Video Processing
Automatic Sign Language Recognition (ASLR) |
- freely available benchmark databases for gesture and sign language recognition
- my diploma thesis Appearance-Based Gesture Recognition (or download it in PS format)
- the new and freely available German fingerspelling alphabet database recorded in the course of my diploma thesis
- the new and freely available RWTH-BOSTON-104 sign language database collected and extended in the course of my current PhD thesis work
- the diploma thesis of David Rybach on the topic Appearance-based Features for Automatic Continuous Sign Language Recognition.
Similar to automatic speech recognition (ASR), we focus in automatic sign language recognition (ASLR) on automatically recognizing sign language videos as gloses, which can be later translated by a statistical machine transaltion system into written text (see a demo video).
Best Student Paper Award: Our work about Speech Recognition Techniques for a Sign Language Recognition System received the ISCA Award for the best student paper of Interspeech 2007. See my publications.
Tracking |
- video clips of dynamic programming tracking under noisy circumstances
- For head tracking on the DUDEK sequence, we use a Viola
and Jones model-based scoring function within our DPT
framework. An example where simply motion is used to track a
hand signing some Palm's Graffiti didgits can be found in the
follwing video clip. We used a partial traceback in order to
reduce the computation power and memory requirements. The
other examples are extracted from the RWTH-BOSTON-104 sign
language database, also in combination with simple motion
based scoring functions.
- we published a paper about Tracking Using Dynamic Programming for Appearance-Based Sign Language Recognition, IEEE FG 2006
- the same tracking framework has been extendend in Pan, Zoom, Scan - Time-coherent, Trained Automatic Video Cropping, IEEE CVPR 2008
Optical Character Recognition (OCR) |
- Currently I am also working on offline recognition of Arabic, Farsi, English, Spanish, and French scripts.
- Our baseline Arabic handwriting recognition system has been presented in White-Space Models for Offline Arabic Handwriting Recognition, ICPR 2008
- Extensions using Writer Adaptive Training (WAT) and modified MMI training for model adaptation have been presented in Writer Adaptive Training and Writing Variant Model Refinement for Offline Arabic Handwriting Recognition and Confidence-Based Discriminative Training for Model Adaptation in Offline Arabic Handwriting Recognition, ICDAR 2009
2D-warping for Face Recognition |
- Face Recognition Using Distortion Models
- Matching Algorithms for Image Recognition
- Our grid-based local feature based face recognition approach has been presented in SURF-Face: Face Recognition Under Viewpoint Consistency Constraints, BMVC 2009
My personal research interests are: |
- sign language recognition
- gesture / fingerspelling recognition
- offline (Arabic) handwriting recognition
- face recognition
- video-analysis and tracking
- complex-scene and image object recognition
- pattern recognition
Philippe Dreuw Last modified: Mon Dec 27 15:25:15 CET 2010 Disclaimer. Created Wed Dec 22 18:04:32 CET 2004