Saliency, Scale and Image Description

@article{Kadir2004SaliencySA,
  title={Saliency, Scale and Image Description},
  author={Timor Kadir and Michael Brady},
  journal={International Journal of Computer Vision},
  year={2004},
  volume={45},
  pages={83-105}
}
  • T. KadirM. Brady
  • Published 1 November 2001
  • Computer Science
  • International Journal of Computer Vision
Many computer vision problems can be considered to consist of two main tasks: the extraction of image content descriptions and their subsequent matching. The appropriate choice of type and level of description is of course task dependent, yet it is generally accepted that the low-level or so called early vision layers in the Human Visual System are context independent.This paper concentrates on the use of low-level approaches for solving computer vision problems and discusses three inter… 

Saliency in images and video: a brief survey

The role and advancement of saliency algorithms over the past decade are surveyed, with an outline of the datasets and performance measures utilised as well as the computational techniques pervasive in the literature.

Salient Regions from Scale-Space Trees

The connection between MSERs and morphological scale-space is demonstrated, which can be enhanced to form a saliency tree which is evaluated via its effectiveness at a standard image retrieval task and which out-performs scale-saliency methods.

Finding Objects of Interest in Images using Saliency and Superpixels

This thesis presents a superpixel segmentation algorithm that outperforms previous algorithms in terms quantitative measures of under-segmentation error and boundary recall, and offers several other advantages over existing algorithms like compactness, uniform size, control on the number of superpixels, and computational efficiency.

Saliency filters: Contrast based filtering for salient region detection

A conceptually clear and intuitive algorithm for contrast-based saliency estimation that outperforms all state-of-the-art approaches and can be formulated in a unified way using high-dimensional Gaussian filters.

Saliency, Scale and Information: Towards a Unifying Theory

Based on the proposed definition of visual saliency, results competitive with the state-of-the art for both prediction of human fixations, and segmentation of salient objects are demonstrated.

Extracting Visual Saliency Based on Multi-scale Receptive Field Template

  • P. Zhang
  • Psychology
    2011 Second International Conference on Digital Manufacturing & Automation
  • 2011
This paper proposes a novel approach to extracting visual saLiency in terms of the statistical features difference between target region and background region and detects the scale of target region adaptively with a set of multi-scale measurements in the multi-resolution framework.

Object component models using Gabor filters for visual recognition

This thesis introduces novel methods using Gabor filters for interest point selection and image region description and shows that approaches based on Gabor filter responses outperform state-of-the-art approaches in several aspects of the object recognition problem.

Contextual information based visual saliency model

This work introduces a general framework for detecting saliency of an image using contextual information and proposes a visual saliency model based on color and shape features that can provide good performance on challenging images including images with cluttered background and repeating distractors compared to the other models.

Occlusion boundaries: low-level detection to high-level reasoning

This thesis focuses on the exploitation of subtle relative-motion cues present at occlusion boundaries, and presents a novel, mid-level model for reasoning more globally about object boundaries and propagating such local information to extract improved, extended boundaries.

A proto-object-based computational model for visual saliency.

This paper employs state-of-the-art computer vision techniques to implement a proto-object-based model for visual attention and evaluates the performance of the proposed method and its components on two challenging eye-fixation datasets.
...

References

SHOWING 1-10 OF 52 REFERENCES

Saliency Maps and Attention Selection in Scale and Spatial Coordinates: An Information Theoretic Approach

A scale space based measure of image information that gives a clear indication of characteristic lengths in a variety of real world images and is superior to power spectrum based measurements is devised.

Scale selection for differential operators

A proper representation of scale is essential to most visual tasks requiring stable descriptors of image structure, and in certain problems, such as shape-from-texture, scale variations in an image also constitute a primary cue in its own right.

Color indexing

It is demonstrated that color histograms of multicolored objects provide a robust, efficient cue for indexing into a large database of models and that they can differentiate among a large number of objects.

A Combined Corner and Edge Detector

The problem we are addressing in Alvey Project MMI149 is that of using computer vision to understand the unconstrained 3D world, in which the viewed scenes will in general contain too wide a

Linear scale-space

The formulation of afront-end or “early vision” system is addressed, and its connection with scale-space is shown, and it is shown that these symmetries suffice to establish the functionality properties of a front-end.

Recovering and characterizing image features using an efficient model based approach

The development of an efficient model-based approach to detect and characterize precisely important features such as edges, corners and vertices is discussed. The key is to propose some efficient

Matching images with different resolutions

This paper shows how to extract interest points at variable scales and devise a method allowing the comparison of two images at two different resolutions, which comprises the use of photometric- and rotation-invariant descriptors, a geometric model mapping the high-resolution image onto a low- resolution image region, and an image matching strategy based on the robust estimation of this geometric model.

Linear Scale-Space I: Basic Theory

This view captures several of the essential aspects of vision as an information processing task, in which an internal representation of information is of out-most importance.

Constructing simple stable descriptions for image partitioning

  • Y. G. Leclerc
  • Computer Science
    International Journal of Computer Vision
  • 2004
A new formulation of the image partitioning problem is presented: construct a complete and stable description of an image-in terms of a specified descriptive language-that is simplest in the sense of being shortest, which yields intuitively satisfying partitions for a wide class of images.

A computational approach for corner and vertex detection

A new scale-space based approach that combines useful properties from the Laplacian and Beaudet's measure (Beaudet 1978) is proposed in order to correct and detect exactly the corner position and an extension of this approach is developed to solve the problem of trihedral vertex characterization and detection.
...