Visual Saliency Detection Based Object Recognition

Below is result for Visual Saliency Detection Based Object Recognition in PDF format. You can download or read online all document for free, but please respect copyrighted ebooks. This site does not host PDF files, all document are the property of their respective owners.

Visual Saliency Prediction Based on Deep Learning

scene, commonly known as a visual saliency detection or visual attention prediction. Recently, neural networks have been used to predict visual saliency. This paper proposes a deep learning encoder-decoder architecture, based on a transfer learning technique, to predict visual saliency. In the

Learning adaptive contrast combinations for visual saliency

[14], image quality assessment [42], and object detection/recognition [2, 16]. The recent years have witnessed great progress in visual saliency detection, and it has received exten-sive attention by the researcher in the fields of psychologists and computer vision [1, 7, 10, 17, 20, 21, 24, 25, 31, 49, 55, 58, 67].

Visual saliency mechanism‐based object recognition with high

sensing object recognition is extracted in this paper. The contrast between the saliency map obtained from this saliency detection method and the original image is shown in Fig. 3. Fig. 1 Ò Framework of visual saliency mechanism-based object recognition with high-resolution remote-sensing images Fig. 2 Ò Framework of computing the saliency map


Saliency detection has been a hot topic in recent years. Its popularity is mainly because of its theoretical meaning for explaining human attention and applicable aims in segmentation, recognition, etc. Nevertheless, traditional algorithms are mostly based on unsupervised techniques, which have limited learning ability.

Object recognition with hierarchical discriminant saliency

attention and recognition. Keywords: object recognition, object detection, top-down saliency, discriminant saliency, hierarchical network. 1. INTRODUCTION. Recent research in computational neuroscience has enabled sig-nificant advances in the modeling of object recognition in visual cortex. These advances are encoded in recent object recogni-

Boosting Bottom-up and Top-down Visual Features for Saliency

distance between posterior and prior beliefs. Graph based Visual Saliency (GBVS) [20] and E-saliency [26]o are tw other methods based on Bayesian and graphical models. Decision theoretic interpretation of saliency states that attention is driven optimality with respect to the end task. Gao and Vasconcelos [35] argued that for recognition,

Visual Saliency Detection Based on Multiscale Deep CNN Features

salient object detection[4], [5]. Since visual saliency results set relative importance on the visual contents in an image, they are conducive to narrowing the scope of visual processing and saving computing resources. As a result, Visual saliency has been incorporated in a variety of computer vision and image

Global Contrast based Salient Region Detection

critical for reliable and coherent saliency detection (see Sec-tion5). 3. Histogram Based Contrast Based on the observation from biological vision that the vision system is sensitive to contrast in visual signal, we propose a histogram-based contrast (HC) method to define saliency values for image pixels using color statistics of the input image.

Saliency-Based Identification and Recognition of Pointed-At

gure-ground segmentation, which is implicit in object-based saliency models. In our experience, this two-phased model, decoupling pointing gesture recognition and identication of the referred-to object, is of advantage, because the object may lie outside the eld of view. Finally, the referred-to object has to be visually focused.

Visual Saliency Detection Based Object Recognition

the object recognition task. Biological visual system tends to nd naturally the most informative regions in a scene. In this paper, we present an object recognition approach based on the visual saliency. Firstly, the most salient region is obtained as the appropriate position of the object with the visual saliency detection method. Secondly

Stereo-Based All-Terrain Obstacle Detection Using Visual Saliency

Stereo-Based All-Terrain Obstacle Detection Using Visual Saliency Pedro Santana LabMAg, University of Lisbon [email protected] Magno Guedes R&D Division, IntRoSys, S.A. Luís Correia LabMAg, University of Lisbon José Barata UNINOVA, New University of Lisbon Last revision: January 4, 2012 Abstract

BiconNet: An Edge-preserved Connectivity-based Approach for

Keywords: Salient object detection, Visual saliency, Connectivity modeling, Deep learning, Edge modeling Corresponding author Email addresses: [email protected] (Ziyun Yang), [email protected] (Somayyeh Soltanian-Zadeh), [email protected] (Sina Farsiu) Preprint submitted to Pattern Recognition June 30, 2021

Saliency-Based Detection for Maritime Object Tracking

This paper presents a new method for object detection and tracking based on visual saliency as a way of mitigat-ing against challenges present in maritime environments. Object detection is based on adaptive hysteresis threshold-ing of a saliency map generated with a modified version of the Boolean Map Saliency (BMS) approach. We show that

Pattern Recognition Letters - GitHub Pages

Visual saliency detection has been applied in many tasks in the fields of pattern recognition and com-puter vision, such as image segmentation, object recognition, and image retargeting. However, the accu-rate detection of saliency remains a challenge. The reasons behind this are that: (1) well-defined

Saliency Detection via Graph-Based Manifold Ranking

The task of saliency detection is to identify the most im-portant and informative part of a scene. It has been applied to numerous vision problems including image segmenta-tion [11], object recognition [28], image compression [16], content based image retrieval [8], to name a few. Saliency methods in general can be categorized as either bottom-up


For this reason, visual saliency detection has been of great research interest [1, 2, 3] in recent years. Analysis of visual attention has benefited a wide range of applications such as object and action recognition, image quality assessment and more. Gao et al. [4] used discrim-inative saliency detection for visual recognition and showed

Black-Box Explanation of Object Detectors via Saliency Maps

more complex detection networks because it does not rely on gradients or the inner workings of the underlying object detector. However, the method in [27] is only applicable to classification, not detection. D-RISE is a black-box method and can be in principle applied to any object detector. Explaining visual classifiers with saliency maps

An efficient visual saliency detection model based on Ripplet

of object-of-interest in an image, automatic image thumb-nailing, object detection and recognition, visual tracking, automatic creation of image collage, content-aware image resizing, non-photo realistic rendering, etc. [2]. Humans exhibit a strong and innate capability to process a visual scene and extract visual saliency features. As far as com-

Visual Saliency Detection Using Group Lasso Regularization in

on only the relevant stimulus. Saliency detection is a chal-lenging problem that has yet to be fully solved. Visual saliency has been the focus of many studies in computer vision in recent years, because of its broad poten-tial applications. Saliency detection can be used for object detection (Navalpakkam and Itti 2006), automatic image

A novel fully convolutional network for visual saliency

saliency prediction, and those models provide good performance. However, those models essentially proposed for object recognition and then fine-tuned for saliency prediction. Consequently, the pixel-based classification of visual attention task remains challenging.

Saliency-based Object Recognition in 3D Data

ble recognition system for the fast detection and classification of objects in spatial 3D data. Depth and reflection data from a 3D laser scanner are rendered into images and fed into a saliency-based visual attention system that detects regions of potential interest. Only these regions are examinated by a fast classifier.

Visual saliency detection with center shift

feature-based visual saliency and location-based visual saliency. After that, we simulate a shifting process of the center of the visual field, which is called center shift, and then present the multiscale analysis. Fig. 1. Different sample images and ground truths for different tasks. (a) Ground truths for salient object detection [14].

Saliency-based object recognition in video

Furthermore visual object recognition in video is an open problem whatever the nature of the content. Incontrast tothewell-known slidingwindow approaches for ob-ject detection and recognition [16, 22], and due to the specific na-ture of the first-person view contents, we aim to drive the object recognition process using visual saliency.

Online Visual Place Recognition Via Saliency Re-Identification

place recognition as saliency re-identication. Both saliency detection and retrieval are performed in frequency domain to take advantage of efciency of element-wise operation. Different from general object detection, saliency detection extracts areas that are visually appealing. This gure shows that the salient regions are re-identied when the

Visual Saliency Detection with Comprehensive Information

Saliency detection for stereoscopic images based on depth confidence analysis and multiple cues fusion Foreground Saliency Using Multiple Cues Contrast When the salient regions have similar appearances with background, the regions may be wrongly detected. Hence, a foreground saliency detection method based on multiple cues contrast is

Category-Independent Object-Level Saliency Detection

3. Saliency Detection with Object-level Infor-mation In this section, we formally describe the algorithm we proposed to perform saliency detection based on high-level object information. 3.1. Object Detection Our method starts with finding an informative prior that captures the potential salient regions from images. While

Visual Saliency Based on Multiscale Deep Features

tecture to make deep CNN features applicable to saliency modeling and salient object detection. 2. Saliency Inference with Deep Features As shown in Fig.1, the architecture of our deep feature based model for visual saliency consists of one output layer and two fully connected hidden layers on top of three deep convolutional neural networks.

Saliency Detection: A Boolean Map Approach

many applications, e.g. image segmentation [12], object recognition [32] and visual tracking [28]. Many previous works have exploited the contrast and the rarity properties of local image patches for saliency detection [19,6,3]. However, these properties have limited ability to model some global perceptual phenomena [23]


Key words to describe this work: Visual attention, video surveillance, region of interest, saliency map, object tracking, motion detection, MPEG video encoder Abstract Object tracking is a very important operation in many surveillance applications. It is also closely related to motion detection/estimation and object recognition.


Attentional system for object detection integrating local saliency and contextual priors about target location. of finding a set of local features in the image. We use a probabilistic definition of saliency that more naturally fits with object detection and recognition formulations we later show : S(x)=p(vl)−1 (1)

BASNet: Boundary-Aware Salient Object Detection

of saliency map and the zoom-in view of boundary map. to traditional methods, their predicted saliency maps are still defective in fine structures and/or boundaries (see Figs. 1(c)-1(d)). There are two main challenges in accurate salient object detection: (i) the saliency is mainly defined over the global

Object Reading: Text Recognition for Object Recognition

Abstract. We propose to use text recognition to aid in visual object class recognition. To this end we rst propose a new algorithm for text detection in natural images. The proposed text detection is based on saliency cues and a context fusion step. The algorithm does not need any parameter tuning and can deal with varying imaging conditions. We

Salient Object Detection by Composition

lem of generic salient object detection. 1. Introduction Humans can identify salient areas in their visual fields with surprising speed and accuracy before performing ac-tual recognition. Simulating such an ability in machine vi-sion is critical and there has been extensive research on this direction [8, 24, 27, 12, 6, 25, 26, 17, 18, 20, 23

Saliency Detection: A Spectral Residual Approach

fast and robust saliency detection of our method. 1. Introduction The first step towards object recognition is object detec-tion. Object detection aims at extracting an object from its background before recognition. But before perform-ing recognitive feature analysis, how can a machine vision system extract the salient regions from an unknown

Visual saliency based global local feature representation for

Traditional visual saliency detection methods are also widely used in the image field. For example, a graph-based visual saliency method [17] forms active maps on feature channels and then normalises them in a salient manner, a context-aware saliency detection method [18] combines context to emphasise the saliency

Pattern Recognition Letters - GitHub Pages

video frame for further object detection, recognition, tracking, etc. With the goal both to achieve a comparable salience de- tection performance of human visual system and to facilitate various saliency-based applications, a rich number of saliency detection methods have been proposed in the past decade

Regional Principal Color Based Saliency Detection

Regional Principal Color Based Saliency Detection Jing Lou, Mingwu Ren*, Huan Wang School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing, Jiangsu, China Abstract Saliency detection is widely used in many visual applications like image segmentation, object recognition and classification.

Frequency-tuned Salient Region Detection

Visual saliency is the perceptual quality that makes an object, person, or pixel stand out relative to its neighbors and thus capture our attention. Visual attention results both from fast, pre-attentive, bottom-up visual saliency of the retinal input, as well as from slower, top-down memory and volition based processing that is task-dependent

Vol. 8, No. 5, 2017 A Bottom-up Approach for Visual Object

a single chip a complete real-time visual saliency applications. However, there is no prior work on parallel implementation of saliency-based bottom-up visual attention model applied to visual object recognition tasks in many-core coarse-grained architecture based parameterisable softcore. The processing re-