A saliency map based on sampling an image into random rectangular regions of interest

Vikram, Tadmeri Narayan; Tscherepanow, Marko; Wrede, Britta

doi:10.1016/j.patcog.2012.02.009

Cited by 110 publications

(77 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Moreover, due to the large difference between optical HSR-RS images and natural images mentioned previously, one cannot directly borrow the conclusions from natural images to use for optical HSR-RS images. In addition, saliency detection [40][41][42][43] has recently become a hot topic in image analysis, and many works have used this as a sampling strategy, e.g., [13,[37][38][39]. However, how to properly use these sampling methods to help the classification task remains a problem.…”

Section: Related Workmentioning

confidence: 99%

“…The sampling strategies adopted in this paper are illustrated in Table 1 ([28], [40][41][42][43]), and each of them will be reviewed in this section. Random sampling is reported to have better performance than the saliency-based sampling in natural image classification when using large numbers of sampled patches [19].…”

Section: Involved Sampling Strategiesmentioning

confidence: 99%

“…In the process of the human visual attention mechanism, the information is selected in a way to reduce the complexity of scene analysis. Over the past decade, saliency detection has attracted considerable attention [40][41][42][43] for analyzing the salient parts of image scenes quickly and accurately. Some work in the scene classification of optical HSR-RS images has attempted to use saliency-based sampling [13], but it is still not clear how it can help the scene classification in other feature learning methods compared to other strategies.…”

Section: Saliency-based Samplingmentioning

confidence: 99%

See 2 more Smart Citations

A Comparative Study of Sampling Analysis in the Scene Classification of Optical High-Spatial Resolution Remote Sensing Imagery

Xia

et al. 2015

Remote Sensing

View full text Add to dashboard Cite

Scene classification, which consists of assigning images with semantic labels by exploiting the local spatial arrangements and structural patterns inside tiled regions, is a key problem in the automatic interpretation of optical high-spatial resolution remote sensing imagery. Many state-of-the-art methods, e.g., the bag-of-visual-words model and its variants, the topic models and unsupervised feature learning-based approaches, share similar procedures: patch sampling, feature learning and classification. Patch sampling is the first and a key procedure, and it has a considerable influence on the results. In the literature, many different sampling strategies have been used, e.g., random sampling and saliency-based sampling. However, the sampling strategy that is most suitable for the scene classification of optical high-spatial resolution remote sensing images remains unclear. In this paper, we comparatively study the effects of different sampling strategies under the scenario of scene classification of optical high-spatial resolution remote sensing images. We divide the existing sampling methods into two types: random sampling and saliency-based sampling. Here, we consider the commonly-used grid sampling to be a specific type of random sampling method, and the saliency-based sampling consists of keypoint-based sampling and salient region-based sampling. To compare their performances, we rely on a standard bag-of-visual-words model to learn the global features for testing because of its simplicity, robustness and efficiency. In addition, we conduct experiments Remote Sens. 2015, 7 14989 using a Fisher kernel framework to validate our conclusions. The experimental results obtained on two commonly-used datasets using different feature learning methods show that random sampling can provide comparable and even better performance than all of the saliency-based strategies.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Involved Sampling Strategiesmentioning

confidence: 99%

Section: Saliency-based Samplingmentioning

confidence: 99%

See 1 more Smart Citation

A Comparative Study of Sampling Analysis in the Scene Classification of Optical High-Spatial Resolution Remote Sensing Imagery

Xia

et al. 2015

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…A fast saliency detection algorithm proposed in [13] considers centre sensitivity, however, achieves low prediction accuracy. The authors of [14] sample the image into rectangular regions of interest and then compute the local saliencies. However, significant computational complexity is a drawback.…”

Section: Introductionmentioning

confidence: 99%

A low complexity visual saliency model based on in-focus regions and centre sensitivity

Chilukamari

Kannangara

Maxwell

2014

2014 IEEE Fourth International Conference on Consumer Electronics Berlin (ICCE-Berlin)

View full text Add to dashboard Cite

“…This sequential processing of pixels or image windows is in contrast to human visual mechanism. The latter incorporates an attention mechanism that helps humans to focus on the most relevant stimuli based on the task at hand [1][2]. Incorporation of similar strategy in computational vision systems, especially for application of robotics, can have many advantages.…”

Section: Introductionmentioning

confidence: 99%

Segmenting Salient Objects in 3D Point Clouds of Indoor Scenes Using Geodesic Distances

Bhatia¹,

Chalup²

2013

JSIP

View full text Add to dashboard Cite

Visual attention mechanisms allow humans to extract relevant and important information from raw input percepts. Many applications in robotics and computer vision have modeled human visual attention mechanisms using a bottomup data centric approach. In contrast, recent studies in cognitive science highlight advantages of a top-down approach to the attention mechanisms, especially in applications involving goal-directed search. In this paper, we propose a topdown approach for extracting salient objects/regions of space. The top-down methodology first isolates different objects in an unorganized point cloud, and compares each object for uniqueness. A measure of saliency using the properties of geodesic distance on the object's surface is defined. Our method works on 3D point cloud data, and identifies salient objects of high curvature and unique silhouette. These being the most unique features of a scene, are robust to clutter, occlusions and view point changes. We provide the details of the proposed method and initial experimental results.

show abstract

A saliency map based on sampling an image into random rectangular regions of interest

Cited by 110 publications

References 25 publications

A Comparative Study of Sampling Analysis in the Scene Classification of Optical High-Spatial Resolution Remote Sensing Imagery

A Comparative Study of Sampling Analysis in the Scene Classification of Optical High-Spatial Resolution Remote Sensing Imagery

A low complexity visual saliency model based on in-focus regions and centre sensitivity

Segmenting Salient Objects in 3D Point Clouds of Indoor Scenes Using Geodesic Distances

Contact Info

Product

Resources

About