2016
DOI: 10.1016/j.jvcir.2016.07.021
|View full text |Cite
|
Sign up to set email alerts
|

Event recognition in photo albums using probabilistic graphical models and feature relevance

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
10
0

Year Published

2018
2018
2024
2024

Publication Types

Select...
5
1
1

Relationship

0
7

Authors

Journals

citations
Cited by 11 publications
(10 citation statements)
references
References 23 publications
0
10
0
Order By: Relevance
“…Accuracy (%) AgS [3] 41.43 ShMM [3] 55.71 Method in [17] 73.41 R-OS-PGM [18] 74.28 Our method (best iteration) 85.0…”
Section: Methodsmentioning
confidence: 99%
See 1 more Smart Citation
“…Accuracy (%) AgS [3] 41.43 ShMM [3] 55.71 Method in [17] 73.41 R-OS-PGM [18] 74.28 Our method (best iteration) 85.0…”
Section: Methodsmentioning
confidence: 99%
“…More recently, in [17], a deep model is trained on images from personal photo collections to capture the co-occurrences, and frequencies of images features. In [18], features extracted through a deep model are used in a graphical model to capture a link among the pictures in a photo album.…”
Section: Related Workmentioning
confidence: 99%
“…An original approach was taken by [6] who casted photo collections as sequential data and treated sub-events as latent variables associated to each image in an Hidden Markov Models and learned them while training the event classifier. More recently, [8] proposed a probabilistic graphical model to predict the event categories of groups of photos, that relies on high-level visual features such as objects and scenes extracted directly from images by employing a deep learning based approach.…”
Section: Event Recognition In Photo Albumsmentioning
confidence: 99%
“…Developing robust and reliable models for the automatic recognition of scenes is of importance in the field of intelligent systems and artificial intelligence since it directly supports real-life applications. For instance, Scene and event recognition has been previously addressed in the literature [1,29]. Scene recognition for robot localization with indoor localization for mobile robots is one of the emerging application scopes of scene recognition [2,5,21].…”
Section: Introductionmentioning
confidence: 99%