Using illumination invariant descriptors for recognition

Healey,; Slater, Mel

doi:10.1109/cvpr.1994.323851

Cited by 12 publications

(10 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…When the database is populated with new images, each of the images is divided into 64 square patches, and each patch is analyzed by six texture models: color histogram, color histogram energy and entropy [38], color histogram invariant features [9], eigenvectors of RGB covariance [40], the multiscale simultaneous autoregressive method [19], and the tree-structured wavelet transform [39]. The user is requested to select patches from one or more images as "positive examples" for a label (e.g., grass, building).…”

Section: The Photobookmentioning

confidence: 99%

Intelligent Image Databases

Gong

1998

View full text Add to dashboard Cite

Section: The Photobookmentioning

confidence: 99%

Intelligent Image Databases

Gong

1998

View full text Add to dashboard Cite

“…Segmentation is achieved by a recursive subdivision of the image and by the analysis of resulting region level statistics of the random vector. This physics based approach has been extended for object recognition using color information [25], [26].…”

Section: Other Imaging Sensorsmentioning

confidence: 99%

Physics-based integration of multiple sensing modalities for scene interpretation

Nandhakumar¹,

Aggarwal

1997

Proc. IEEE

View full text Add to dashboard Cite

The fusion of multiple imaging modalities offers many advantages over the analysis, separately, of the individual sensory modalities. In this paper we present a unique approach to the integrated analysis of disparate sources of imagery for object recognition. The approach is based on physics-based modeling of the image generation mechanisms. Such models make possible features that are physically meaningful and have an improved capacity to differentiate between multiple classes of objects. We illustrate the use of physics-based approach to develop multisensory vision systems for different object recognition application domains. The paper discusses the integration of different suites of sensors, the integration of image-derived information with model-derived information, and the physics-based simulation of multisensory imagery.

show abstract

“…Other feature-based image retrieval systems can be found in Hirata and Kato (1994), Picard and Minka (1995), Sclaroff and Pentland (1995), Gudivada and Raghavan (1995), Tan and Kittler (1993), Healey and Slater (1994), Mao and Jain (1992), Aigrain et al (1996) and Smeulders et al (2000).…”

Section: Feature-based Methodsmentioning

confidence: 99%

A new strategy for bridging the semantic gap in image retrieval

Alzubaidi

2017

IJCSE

View full text Add to dashboard Cite

Content-based image retrieval (CBIR) research is currently faced with the so called the 'semantic gap' problem. CBIR researchers work at the near end of the gap, applying computer science methods to bridge the gap. Cognitive psychology researchers work at the far end of the gap, studying how humans perceive things in their environment. This article looks at the literature from a new perspective, and presents a structured analysis of what CBIR researchers have done, in their efforts to bridge the gap. This analysis suggests that more emphasis should be placed on studying the far end of the gap. The second part of this article discusses a study conducted at the far end of the gap, to begin the process of developing and testing techniques for externalising and analysing the visual concepts that are evoked by images. This more disciplined approach has the potential to guide researchers in their efforts to bridge the gap.Keywords: image indexing; image retrieval; semantic gap; semantic basis functions; SBFs; affective basis functions; ABFs; basic research; applied research.Reference to this paper should be made as follows: Alzubaidi, M.A. (2017) 'A new strategy for bridging the semantic gap in image retrieval', Int.

show abstract

Using illumination invariant descriptors for recognition

Cited by 12 publications

References 6 publications

Intelligent Image Databases

Intelligent Image Databases

Physics-based integration of multiple sensing modalities for scene interpretation

A new strategy for bridging the semantic gap in image retrieval

Contact Info

Product

Resources

About