Improving the convergence of co-training for audio-visual person identification.
Nicolai Bæk ThomsenXiaodong DuanZheng-Hua TanBørge LindbergSøren Holdt JensenPublished in: SPLINE (2016)
Keyphrases
- audio visual
- co training
- person identification
- multi modal
- semi supervised learning
- semi supervised
- multi view
- text classification
- visual information
- unlabeled data
- supervised learning
- emotion recognition
- multimedia
- gait recognition
- named entities
- face recognition
- visual data
- audio features
- speaker verification
- biometric systems
- active learning
- image data
- data sets
- palmprint
- computer vision
- machine learning