Fingerprint information maximization for content identification
| Title | Fingerprint information maximization for content identification |
| Publication Type | Conference Paper |
| Year of Publication | 2014 |
| Authors | Naini, R., Moulin, P. |
| Conference Name | Acoustics, Speech and Signal Processing (ICASSP), 2014 IEEE International Conference on |
| Date Published | May |
| Keywords | Approximation methods, Audio fingerprinting, content fingerprint, Content Identification, Databases, Dictionaries, distortion channel, filter optimization, fingerprint information maximization, greedy optimization, Information bottleneck, information bottleneck method, Information maximization, information retrieval, Joints, learning based fingerprint, Mutual information, mutual information across, optimisation, Optimization, Quantization (signal), quantizer optimization |
| Abstract | This paper presents a novel design of content fingerprints based on maximization of the mutual information across the distortion channel. We use the information bottleneck method to optimize the filters and quantizers that generate these fingerprints. A greedy optimization scheme is used to select filters from a dictionary and allocate fingerprint bits. We test the performance of this method for audio fingerprinting and show substantial improvements over existing learning based fingerprints. |
| DOI | 10.1109/ICASSP.2014.6854314 |
| Citation Key | 6854314 |
- information bottleneck method
- quantizer optimization
- Quantization (signal)
- optimization
- optimisation
- mutual information across
- Mutual information
- learning based fingerprint
- Joints
- information retrieval
- Information maximization
- Approximation methods
- Information bottleneck
- greedy optimization
- fingerprint information maximization
- filter optimization
- distortion channel
- Dictionaries
- Databases
- Content Identification
- content fingerprint
- Audio fingerprinting