Neural dynamics of surface perception: boundary webs, illuminants, and shape-from-shading
Computer Vision, Graphics, and Image Processing
3D shape representation by contours
Computer Vision, Graphics, and Image Processing
Trace Inference, Curvature Consistency, and Curve Detection
IEEE Transactions on Pattern Analysis and Machine Intelligence
Integrating Region Growing and Edge Detection
IEEE Transactions on Pattern Analysis and Machine Intelligence
Using Dynamic Programming for Solving Variational Problems in Vision
IEEE Transactions on Pattern Analysis and Machine Intelligence
On minimal energy trajectories
Computer Vision, Graphics, and Image Processing
Grouping contours by iterated pairing network
NIPS-3 Proceedings of the 1990 conference on Advances in neural information processing systems 3
Perceptual Organization for Scene Segmentation and Description
IEEE Transactions on Pattern Analysis and Machine Intelligence
Object discrimination based on depth-from-occlusion
Neural Computation
Steerable filters and local analysis of image structure
Steerable filters and local analysis of image structure
Dynamic Programming for Detecting, Tracking, and Matching Deformable Contours
IEEE Transactions on Pattern Analysis and Machine Intelligence
Perceptual completion of occluded surfaces
Computer Vision and Image Understanding
ACM Transactions on Mathematical Software (TOMS)
An application of heuristic search methods to edge and contour detection
Communications of the ACM
On the optimal detection of curves in noisy pictures
Communications of the ACM
Figure-Ground Discrimination: A Combinatorial Optimization Approach
IEEE Transactions on Pattern Analysis and Machine Intelligence
Extracting Salient Curves from Images: An Analysis of the Saliency Network
CVPR '96 Proceedings of the 1996 Conference on Computer Vision and Pattern Recognition (CVPR '96)
Stochastic completion fields: a neural model of illusory contour shape and salience
ICCV '95 Proceedings of the Fifth International Conference on Computer Vision
Recognizing 3-D Objects Using 2-D Images
Recognizing 3-D Objects Using 2-D Images
Curved Inertia Frames Visual Attention and Perceptual Organization Using Convexity and Symmetry
Curved Inertia Frames Visual Attention and Perceptual Organization Using Convexity and Symmetry
IEEE Transactions on Pattern Analysis and Machine Intelligence
A Probabilistic Interpretation of the Saliency Network
ECCV '00 Proceedings of the 6th European Conference on Computer Vision-Part II
Query definition using interactive saliency
MIR '03 Proceedings of the 5th ACM SIGMM international workshop on Multimedia information retrieval
On the Distribution of Saliency
IEEE Transactions on Pattern Analysis and Machine Intelligence
Global Detection of Salient Convex Boundaries
International Journal of Computer Vision
GPSel: A Gestural Perceptual-Based Path Selection Technique
SG '09 Proceedings of the 10th International Symposium on Smart Graphics
Object of interest detection by saliency learning
ECCV'10 Proceedings of the 11th European conference on Computer vision: Part II
On the distribution of saliency
CVPR'04 Proceedings of the 2004 IEEE computer society conference on Computer vision and pattern recognition
Contour grouping: focusing on image patches around edges
VSMM'06 Proceedings of the 12th international conference on Interactive Technologies and Sociotechnical Systems
Extraction of salient contours in color images
SETN'06 Proceedings of the 4th Helenic conference on Advances in Artificial Intelligence
On similarity-invariant fairness measures
Scale-Space'05 Proceedings of the 5th international conference on Scale Space and PDE Methods in Computer Vision
| Hi-index | 0.00 |
The Saliency Network proposed by Shashua and Ullman (1988) is awell-known approach to the problem of extracting salient curves fromimages while performing gap completion. This paper analyzes theSaliency Network. The Saliency Network is attractive for severalreasons. First, the network generally prefers long and smooth curvesover short or wiggly ones. While computing saliencies, the networkalso fills in gaps with smooth completions and tolerates noise.Finally, the network is locally connected, and its size isproportional to the size of the image.Nevertheless, our analysis reveals certain weaknesses with the method.In particular, we show cases in which the most salient element doesnot lie on the perceptually most salient curve. Furthermore, in somecases the saliency measure changes its preferences when curves arescaled uniformly. Also, we show that for certain fragmented curvesthe measure prefers large gaps over a few small gaps of the same totalsize. In addition, we analyze the time complexity required by themethod. We show that the number of steps required for convergence inserial implementations is quadratic in the size of the network, and inparallel implementations is linear in the size of the network. Wediscuss problems due to coarse sampling of the range of possibleorientations. Finally, we consider the possibility of using theSaliency Network for grouping. We show that the Saliency Networkrecovers the most salient curve efficiently, but it has problems withidentifying any salient curve other than the most salient one.