A Biologically-Inspired Framework for Contour Detection Using Superpixel-Based Candidates and Hierarchical Visual Cues

Sun, Xiaoyang; Shang, Ke; Ming, Delie; Tian, Jinwen; Ma, Jiayi

doi:10.3390/s151026654

Cited by 14 publications

(7 citation statements)

References 72 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Green symbols in A show average y value for individual abscissa values; symbol size scales with number of data points. D shows correlation values for scatter plots in A - C and those generated by other computer vision algorithms (Itti-Koch [ 3 ], GBVS [ 41 ], gPb-HS [ 5 ], nCuts [ 42 ], HVC [ 43 ]); open green symbol plots correlation for top-down map when consensus probe locations (indicated by solid green symbol in A ) are excluded. Error bars in D show 95% confidence intervals.…”

Section: Resultsmentioning

confidence: 99%

“…We applied the following 6 computer vision algorithms from published literature: visual saliency [ 3 ] (Itti-Koch), graph-based visual saliency [ 41 ] (GBVS), hierarchical segmentation [ 5 ] (gPb-HS), normalized cuts [ 42 ] (nCuts), contour detection using superpixel-based candidates and hierarchical visual cues [ 43 ] (HVC), conditional random fields as recurrent neural networks [ 23 ] (CRF-RNN). HVC and gPb-HS had already been applied to BSD500; for those, we obtained results from the algorithm creators.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Object segmentation controls image reconstruction from natural scenes

Neri

2017

PLoS Biol

View full text Add to dashboard Cite

The structure of the physical world projects images onto our eyes. However, those images are often poorly representative of environmental structure: well-defined boundaries within the eye may correspond to irrelevant features of the physical world, while critical features of the physical world may be nearly invisible at the retinal projection. The challenge for the visual cortex is to sort these two types of features according to their utility in ultimately reconstructing percepts and interpreting the constituents of the scene. We describe a novel paradigm that enabled us to selectively evaluate the relative role played by these two feature classes in signal reconstruction from corrupted images. Our measurements demonstrate that this process is quickly dominated by the inferred structure of the environment, and only minimally controlled by variations of raw image content. The inferential mechanism is spatially global and its impact on early visual cortex is fast. Furthermore, it retunes local visual processing for more efficient feature extraction without altering the intrinsic transduction noise. The basic properties of this process can be partially captured by a combination of small-scale circuit models and large-scale network architectures. Taken together, our results challenge compartmentalized notions of bottom-up/top-down perception and suggest instead that these two modes are best viewed as an integrated perceptual mechanism.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Object segmentation controls image reconstruction from natural scenes

Neri

2017

PLoS Biol

View full text Add to dashboard Cite

show abstract

“…Martin et al [ 6 ] proposed the Pb method by defining gradient operators for brightness, colour, and texture channels, using them as input to a logistic regression classifier for predicting edge strength, and finally integrating the predicted edges with supervised learning method. In subsequent studies, researchers further improved the effect of contour detection mainly by introducing technologies, such as multi-feature [ 10 ], multi-scale [ 11 ] and global information [ 12 ]. However, this supervised method relies too much on the training sets and has low computational efficiency.…”

Section: Introductionmentioning

confidence: 99%

Biologically Inspired Hierarchical Contour Detection with Surround Modulation and Neural Connection

Cong

et al. 2018

Sensors

View full text Add to dashboard Cite

Contour is a very important feature in biological visual cognition and has been extensively investigated as a fundamental vision problem. In connection with the limitations of conventional models in detecting image contours in complex scenes, a hierarchical image contour extraction method is proposed based on the biological vision mechanism that draws on the perceptual characteristics of the early vision for features such as edges, shapes, and colours. By simulating the information processing mechanisms of the cells’ receptive fields in the early stages of the biological visual system, we put forward a computational model that combines feedforward, lateral, and feedback neural connections to decode and obtain the image contours. Our model simulations and their results show that the established hierarchical contour detection model can adequately fit the characteristics of the biological experiment, quickly and effectively detect the salient contours in complex scenes, and better suppress the unwanted textures.

show abstract

“…The algorithm may locally manage the calculation of each scale by ordering by the edges with the edges. The visual system based on cue combination and contextual modulation in multifeature-based surround inhibition (MFBSI) method [11]. The local features contribution is also considered in cluttered natural scenes for contour identification.…”

Section: Introductionmentioning

confidence: 99%

A Multiscale Particle Filter and Multifeature Based Contour Detection

Swami¹,

Chaurasia²,

Kurmi³

2017

ijoaem

View full text Add to dashboard Cite

In this paper, a multi-scale particle filter (MSPF) based algorithm and multi-feature-based surround inhibition (MFBSI) algorithm are proposed for contour detection in natural images. The contour detection algorithms are jointly tracks at two edgelets as edge elements. This multi-scale edgelet structure gathers the semi-local information as Bayesian modelling. The model is further approximate by using Monte Carlo algorithm. The individual surround inhibition weights of the feature as, including luminance, luminance contrast, and orientation. The scale-guided strategy is used for features together. The final surround inhibition is modulated using the combined weights in the neuron. The luminance contrast and luminance are proven as an excellent contour extraction capability. The multiscale particle filter (MSPF) based approach provides a better contour for the object segmentation.

show abstract

A Biologically-Inspired Framework for Contour Detection Using Superpixel-Based Candidates and Hierarchical Visual Cues

Cited by 14 publications

References 72 publications

Object segmentation controls image reconstruction from natural scenes

Object segmentation controls image reconstruction from natural scenes

Biologically Inspired Hierarchical Contour Detection with Surround Modulation and Neural Connection

A Multiscale Particle Filter and Multifeature Based Contour Detection

Contact Info

Product

Resources

About