An experimental study on the universality of visual vocabularies

Abstract and Applied Analysis

Karimi

2014

Self Cite

Feature combination is a powerful approach to improve object classification performance. While various combination algorithms have been proposed, average combination is almost always selected as the baseline algorithm to be compared with. In previous work we have found that it is better to use only a sample of the most powerful features in average combination than using all. In this paper, we continue this work and further show that the behaviors of features in average combination can be integrated into thek-Nearest-Neighbor (kNN) framework. Based on the kNN framework, we then propose to use a selection based average combination algorithm to obtain the best classification performance from average combination. Our experiments on four diverse datasets indicate that this selection based average combination performs evidently better than the ordinary average combination, and thus serves as a better baseline. Comparing with this new and better baseline makes the claimed superiority of newly proposed combination algorithms more convincing. Furthermore, the kNN framework is helpful in understanding the underlying mechanism of feature combination and motivating novel feature combination algorithms.

Section: Knn Frameworkmentioning

confidence: 99%

Exploring the Best Classification from Average Feature Combination

Abstract and Applied Analysis

Karimi

2014

Self Cite

“…Furthermore, the universal vocabulary obtained with our approach is optimal and compact, in that it can be used on different datasets to obtain the (near-)best performance, and the vocabulary size is only several thousands. These two important properties are not possessed by the universal vocabularies achieved in [14,15].…”

Section: Introductionmentioning

confidence: 94%

“…In the literature, the most related works to ours are the ones in [14,15] which address the problem of deriving a universal visual vocabulary. Specifically, it was empirically found in [14,15] that the visual vocabularies trained from one dataset can be used on some other datasets without apparently harming the performance, only if the dataset is large enough. In these two papers, the vocabulary sizes are still user-defined and this implies that an inappropriate vocabulary size may lead to a universal vocabulary that performs moderately, which obviously is not what we really expect.…”

Section: Introductionmentioning

confidence: 99%

On Building a Universal and Compact Visual Vocabulary

Mathematical Problems in Engineering

et al. 2013

Self Cite

Bag-of-visual-words has been shown to be a powerful image representation and attained great success in many computer vision and pattern recognition applications. Usually, for a given dataset, researchers choose to build a specific visual vocabulary from the dataset, and the problem of deriving a universal visual vocabulary is rarely addressed. Based on previous work on the classification performance with respect to visual vocabulary sizes, we arrive at a hypothesis that a universal visual vocabulary can be obtained by taking-into account the similarity extent of keypoints represented by one visual word. We then propose to use a similarity threshold-based clustering method to calculate the optimal vocabulary size, where the universal similarity threshold can be obtained empirically. With the optimal vocabulary size, the optimal visual vocabularies of limited sizes from three datasets are shown to be exchangeable and therefore universal. This result indicates that a universal and compact visual vocabulary can be built from a not too small dataset. Our work narrows the gab between bag-of-visual-words and bag-of-words, where a relatively fixed vocabulary can be used with different text datasets.

“…They build a codebook for a given database, and the codebook is not used in other image databases. The literatures [26][27][28][29] have explored the possibility of building the universal codebook, which can be used in various image databases.…”

Section: Related Workmentioning

confidence: 99%

Research on Vocabulary Sizes and Codebook Universality

Abstract and Applied Analysis

Karimi

2014

Self Cite

Codebook is an effective image representation method. By clustering in local image descriptors, a codebook is shown to be a distinctive image feature and widely applied in object classification. In almost all existing works on codebooks, the building of the visual vocabulary follows a basic routine, that is, extracting local image descriptors and clustering with a user-designated number of clusters. The problem with this routine lies in that building a codebook for each single dataset is not efficient. In order to deal with this problem, we investigate the influence of vocabulary sizes on classification performance and vocabulary universality with the kNN classifier. Experimental results indicate that, under the condition that the vocabulary size is large enough, the vocabularies built from different datasets are exchangeable and universal.