2012
DOI: 10.1007/978-3-642-34166-3_19
|View full text |Cite
|
Sign up to set email alerts
|

A Relational Kernel-Based Framework for Hierarchical Image Understanding

Abstract: While relational representations have been popular in early work on syntactic and structural pattern recognition, they are rarely used in contemporary approaches to computer vision due to their pure symbolic nature. The recent progress and successes in combining statistical learning principles with relational representations motivates us to reinvestigate the use of such representations. More specifically, we show that statistical relational learning can be successfully used for hierarchical image understanding… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
13
0

Year Published

2013
2013
2020
2020

Publication Types

Select...
5
2
1

Relationship

4
4

Authors

Journals

citations
Cited by 13 publications
(13 citation statements)
references
References 21 publications
0
13
0
Order By: Relevance
“…Examples of sequence of the eyes gray level characteristics curves are shown in Figure 9, and the results of person recognition through the eyes gray level characteristics are shown in Table I. Table I shows comparison among persons characteristics in calculation of standard deviation values of pixel gray level differences in equation (8) and average values of negative and positive gradients counting differences in equation (12). Table 1 shows comparison among three persons characteristics in many different poses, with eyeglasses or not.…”
Section: Combination Of Face and Posture Features For Tracking Of Movingmentioning
confidence: 99%
See 1 more Smart Citation
“…Examples of sequence of the eyes gray level characteristics curves are shown in Figure 9, and the results of person recognition through the eyes gray level characteristics are shown in Table I. Table I shows comparison among persons characteristics in calculation of standard deviation values of pixel gray level differences in equation (8) and average values of negative and positive gradients counting differences in equation (12). Table 1 shows comparison among three persons characteristics in many different poses, with eyeglasses or not.…”
Section: Combination Of Face and Posture Features For Tracking Of Movingmentioning
confidence: 99%
“…Besides statistical methods, syntactical methods include hierarchical, relational, structural, and morphological methods have been developed for face recognition [7][8][9][10]. Structural hierarchical and relational methods have been done for high level abstraction of image objects modelling, but still have problems for low level implementation [7][8].…”
Section: Introductionmentioning
confidence: 99%
“…Our method outperforms the baselines and the boosting approach. Results on 60 images with hierarchy -houses In previous work [49], we performed the same experiments on a subset of 60 images of the dataset considered in this paper. For comparison, the results are summarized in Table 3.…”
Section: Application and Experimental Evaluationmentioning
confidence: 99%
“…Many statistical and machine learning methods need huge amount or training data taken from human perception or knowledge about objects [10], but it is still incomplete collected training data [11]. Besides statistical methods, syntactical methods include hierarchical, relational, structural, and morphological methods have been developed for face recognition [12][13][14] [15]. Structural hierarchical and relational methods have been done for high level abstraction of image objects modelling, but still have problems for low level implementation [12] [13].…”
Section: Introductionmentioning
confidence: 99%
“…Besides statistical methods, syntactical methods include hierarchical, relational, structural, and morphological methods have been developed for face recognition [12][13][14] [15]. Structural hierarchical and relational methods have been done for high level abstraction of image objects modelling, but still have problems for low level implementation [12] [13]. Many researchers have developed various human activities and behaviour recognition, such as walking, sitting, bending, and some sport activities, and less work of person identification that does the activities [15].…”
Section: Introductionmentioning
confidence: 99%