Robust Visual Vocabulary Based On Grid Clustering
| Autor: | Michel Dhome, Marc Chevaldonné, Eric Royer, Achref Ouni |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
Vocabulary
Phrase Contextual image classification business.industry Computer science media_common.quotation_subject InformationSystems_INFORMATIONSTORAGEANDRETRIEVAL ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Pattern recognition Grid clustering ComputingMethodologies_PATTERNRECOGNITION Discriminative model Image representation Bag-of-words model in computer vision Artificial intelligence business Image retrieval media_common |
| Zdroj: | Intelligent Decision Technologies ISBN: 9789811627644 KES-IDT |
| DOI: | 10.1007/978-981-16-2765-1_18 |
| Popis: | Content-based image retrieval (CBIR) is the task of finding the images in the dataset that are considered similar to an input query based on their visual content. Many methods based on visual description try to solve the CBIR problem. In particular, bag of visual words (BoVW) is one of the most algorithm used to image classification and recognition. But, even with the discriminative power of BoVW, this problem is still a challenge in computer vision. We propose in this paper an efficient CBIR approach based on bag of visual words model (BoVW). Our aim here is to improve the image representation by transforming the BoVW model to the bag of visual phrase (BoVP) based on grid clustering approach. We show experimentally that the proposed model leads to the increase of accuracy of CBIR results. We study the performance of the proposed approach on four different datasets (Corel 1 K, UKB, Holidays, MSRC v1) and two descriptors (SURF, KAZE). |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|