Softening quantization in bag-of-audio-words
Authors:
Open Access
- 1 May 2014
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15206149,p. 1370-1374
Abstract
The audio component of multimedia data can be crucial for multimedia content analysis. Bag-of-audio-words (BoAW) approach is one of the most frequently used methods to represent audio content in multimedia event detection and related tasks. The method, however, has numerous criticisms, amongst which is the loss of information in the “vector quantization” step which generates word-like units. In this work, we address this issue by employing a soft quantization representation where the distance to the nearest codeword is incorporated into the model, rather than only using the nearest codeword's index as is the case with hard quantization. We explore two techniques for soft quantization and apply it to the BoAW for multimedia event detection. We find the best setup yields a 13% improvement in mean average precision, improving performance for 27 of the 30 video events.Keywords
This publication has 6 references indexed in Scilit:
- Supervised acoustic concept extraction for multimedia event detection2012
- Learning mid-level features for recognition2010
- Visual Word AmbiguityIEEE Transactions on Pattern Analysis and Machine Intelligence, 2009
- Linear spatial pyramid matching using sparse coding for image classification2009
- Large-scale content-based audio retrieval from text queries2008
- Lost in quantization: Improving particular object retrieval in large scale image databases2008