Noise-robust digit recognition with exemplar-based sparse representations of variable length.
Emre YilmazJort F. GemmekeDirk Van CompernolleHugo Van hammePublished in: MLSP (2012)
Keyphrases
- variable length
- digit recognition
- sparse representation
- noisy environments
- fixed length
- dictionary learning
- noise reduction
- sparse coding
- image representation
- image classification
- signal processing
- image patches
- hand written
- joint optimization
- bitstream
- speech recognition
- dimensionality reduction
- data sets
- n gram
- face recognition
- signal to noise ratio
- visual information
- human motion
- high dimensional data
- automatic speech recognition
- visual features
- data analysis
- three dimensional