Bags of phrases with codebooks alignment for near duplicate image detection

Battiato, Sebastiano; Farinella, Giovanni Maria; Guarnera, Giuseppe Claudio; Meccio, Tony; Puglisi, Giovanni; Rizzo, Rosetta

doi:10.1145/1877972.1877991

Cited by 14 publications

(11 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The first kind of methods are designed for single-media (image [17], [18]) or multi-modal (image and text [19], [20], [21], image and audio [22]) applications. The second kind of methods can learn cross-media correlations among different media types.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Semi-Supervised Cross-Media Feature Learning With Unified Patch Graph Regularization

Peng

Zhai

Zhao

et al. 2016

IEEE Trans. Circuits Syst. Video Technol.

133

View full text Add to dashboard Cite

With the rapid growth of multimedia data such as text, image, video, audio and 3D model, cross-media retrieval has become increasingly important, because users can retrieve the results with various types of media by submitting a query of any media type. Comparing with single-media retrieval such as image retrieval and text retrieval, cross-media retrieval is better because it provides the retrieval results with all kinds of media at the same time. In this paper, we focus on how to learn cross-media features for different media types, which is a key challenge for cross-media retrieval. Existing methods either model different media types separately or only exploit the labeled multimedia data. Actually, data from different media types with the same semantic category are complementary to each other, and jointly modeling them is able to improve the accuracy of cross-media retrieval. In addition, although the labeled data are accurate, they require a lot of human labor and thus are very scarce. To address the above problems, we propose a semi-supervised crossmedia feature learning algorithm with unified patch graph regularization (S 2 UPG). Our motivation and contributions mainly lie in the following three aspects: (1) Existing methods only model different media types in different graphs, while we employ one joint graph to simultaneously model all the media types. The joint graph is able to fully exploit the semantic correlations among various media types, which are complementary to provide the rich hint for cross-media correlation. (2) Existing methods only consider the original media instances (such as images, videos, texts, audios, and 3D models) but ignore their patches, while we make full use of both the media instances and their patches in one graph. Cross-media patches could emphasize the important parts and make cross-media correlations more precise.(3) Traditional semi-supervised learning methods only exploit single-media unlabeled instances, while our approach fully exploits cross-media unlabeled instances and their patches, which can increase the diversity of training data and boost the accuracy of cross-media retrieval. Comparing with the current state-of-theart methods on 3 datasets, including the challenging XMedia dataset with 5 media types, the comprehensive experimental results show that our proposed approach performs better.

show abstract

Section: Related Workmentioning

confidence: 99%

“…This method is further improved by Battiato et al [18]. They find that there exists coherence between spaces in both the image representation and the codebook steps.…”

Section: Related Workmentioning

confidence: 99%

Semi-Supervised Cross-Media Feature Learning With Unified Patch Graph Regularization

Peng

Zhai

Zhao

et al. 2016

IEEE Trans. Circuits Syst. Video Technol.

133

View full text Add to dashboard Cite

show abstract

“…Para realizar a classificação das imagens eles utilizaram a distribuição aprendida a partir do conjunto de treinamento representando características de co-ocorrência de cor e textura dos textons por meio de um histograma, diferentemente dos trabalhos citados anteriormente, que utilizam os centróides dos grupos encontrados. A análise de textons por histograma (BATTIATO et al, 2008;SHOTTON;CIPOLLA, 2008;FEI-FEI;PERONA, 2005) representa uma imagem pela frequência de textons sem levar em consideração onde as características estão localizadas, mas dando importância à sua frequência, de maneira similar a estatísticas de primeira ordem.…”

Section: Textonsunclassified

“…As palavras visuais são vetores representativos resultantes de um procedimento de agrupamento dos vetores de características que representam cada região. Em geral, o algoritmo k-means é utilizado para realizar o agrupamento e diversos trabalhos utilizam os centróides de cada grupo para representação dos padrões (BATTIATO et al, 2008;BATTIATO et al, 2010;GALARO et al, 2011;CROSIER;GRIFFIN, 2010). No entanto, alguns trabalhos utilizam uma amostra aleatória de cada grupo afirmando que não há um impacto significante na classificação de texturas naturais quando um algoritmo de agrupamento é utilizado (NOWAK; JURIE; TRIGGS, 2006).…”

Section: Bag-of-wordsunclassified

“…Quando faz-se necessário analisar a posição das palavras visuais, como por exemplo para a caracterização de cenas, são usadas abordagens hierárquicas (LAZEBNIK; SCHMID; PONCE, 2006;BATTIATO et al, 2008). De acordo com trabalhos de caracterização de texturas naturais, quanto maior o tamanho do vocabulário melhor é a identificação (NOWAK; JURIE; TRIGGS, 2006;CSURKA et al, 2004).…”

Section: Bag-of-wordsunclassified

See 1 more Smart Citation