Histology Image Indexing Using a Non-negative Semantic Embedding

Vanegas, Jorge A.; Caicedo, Juan C.; Gonz, Fabio A.; lez,; Romero, Eduardo

doi:10.1007/978-3-642-28460-1_8

Cited by 4 publications

(17 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Chandrika et al [9] evaluated a multimodal Latent Semantic Indexing (mmLSI) algorithm, based on Singular Value Decomposition (SVD), and Caicedo et al [6] and Akata et al [1] proposed finding multimodal relationships using Nonnegative Matrix Factorization (NMF). Vanegas et al [29] proposed the Non-negative Semantic Embedding (NSE) as a variant on the NMF algorithm, where the semantic encoding is already known and is defined by the text modality, and the problem is simplified to find a transformation matrix which represents the relationships between the visual and text content. The experimental evaluation in [29] showed that NSE performs better than other semantic embedding models when the text modality corresponds to clean and semantic rich annotations.…”

Section: Related Workmentioning

confidence: 99%

“…Vanegas et al [29] proposed the Non-negative Semantic Embedding (NSE) as a variant on the NMF algorithm, where the semantic encoding is already known and is defined by the text modality, and the problem is simplified to find a transformation matrix which represents the relationships between the visual and text content. The experimental evaluation in [29] showed that NSE performs better than other semantic embedding models when the text modality corresponds to clean and semantic rich annotations.…”

Section: Related Workmentioning

confidence: 99%

“…The algorithm presented in this paper, called ONSE is a reformulation of the classic NSE [29] as an online-learning algorithm using a stochastic gradient descent approach, which only needs to keep small portions of the training data in main memory and requires very little time to reach a predefined expected risk. ONSE retains the advantages of NSE which make it perform better than other matrix factorization approaches when applied to images with clean and semantic-rich annotations.…”

Section: S103mentioning

confidence: 99%

“…This strategy has been previously approached using different methods: Latent Semantic Analysis (LSA) [21], Latent Dirichlet Allocation (LDA) [21], Non-negative Matrix Factorization (NMF) [6] and Non-negative Semantic Embedding (NSE) [29], among others. Unfortunately most semantic learning algorithms demand huge computational resources in terms of memory and time processing [9], making them infeasible for large data collections.…”

Section: Introductionmentioning

confidence: 99%

“…To solve this drawback, this paper proposes an online version of the NSE algorithm [29], online nonnegative semantic embedding (ONSE), which is able to deal with large image collections by keeping only a small subset of the training data in main memory. This paper is an extended version of [30] where ONSE was introduced for the first time.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Image indexing using regularized online non-negative semantic embedding

Vanegas

González

2014

IDA

Self Cite

View full text Add to dashboard Cite

This paper proposes a novel method for image indexing based on an online learning approach which can deal with large repositories of images. The proposed method is based on a semantic embedding strategy which models a mapping between visual and text representations. This method enhances the image representation by taking advantage of the text annotations associated to the images, which have a rich and clean semantic interpretation. Once the mapping is learned, a new (annotated or unannotated) image can be projected to the space defined by semantic annotations. In this manner, this method can be used to search into the collection using an image as query (query-by-example strategy) and to annotate new unannotated images. The main drawback of semantic embedding strategies is that the associated algorithms are computationally expensive, making them infeasible for large data collections. In order to address this issue, the proposed method is formulated as an online learning algorithm using the stochastic gradient descent approach, which can scale to deal with large image collections. According with the experimental evaluation, the proposed method, in comparison with several baseline methods, is faster and consumes less memory, without degradation in the performance in content-based image search.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: S103mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Image indexing using regularized online non-negative semantic embedding

Vanegas

González

2014

IDA

Self Cite

View full text Add to dashboard Cite

show abstract

Large Scale Image Indexing Using Online Non-negative Semantic Embedding

Vanegas

González

2013

Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications

Self Cite

View full text Add to dashboard Cite

This paper presents a novel method to address the problem of indexing a large set of images taking advantage of associated multimodal content such as text or tags. The method finds relationships between the visual and text modalities enriching the image content representation to improve the performance of content-based image search.This method finds a mapping that connects visual and text information that allows to project new (annotated and unannotated) images to the space defined by semantic annotations, this new representation can be used to search into the collection using a query-by-example strategy and to annotate new unannotated images. The principal advantage of the proposed method is its formulation as an online learning algorithm, which can scale to deal with large image collections. The experimental evaluation shows that the proposed method, in comparison with several baseline methods, is faster and consumes less memory, keeping a competitive performance in content-based image search.

show abstract