2014 International Joint Conference on Neural Networks (IJCNN) 2014
DOI: 10.1109/ijcnn.2014.6889620
|View full text |Cite
|
Sign up to set email alerts
|

Micro-expression recognition based on local binary patterns from three orthogonal planes and nearest neighbor method

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
13
0

Year Published

2016
2016
2023
2023

Publication Types

Select...
5
3

Relationship

0
8

Authors

Journals

citations
Cited by 33 publications
(16 citation statements)
references
References 12 publications
0
13
0
Order By: Relevance
“…Among appearance-based feature extraction methods, local binary pattern on three orthogonal planes (LBP-TOP) is widely applied in many works (Li et al, 2013; Guo et al, 2014; Le Ngo et al, 2014, 2015, 2016a,b; Yan et al, 2014a; Adegun and Vadapalli, 2016; Zheng et al, 2016; Wang et al, 2017). Most existing datasets (SMIC, CASME II, SAMM) have all reported the LBP-TOP as their baseline evaluation method.…”
Section: Recognition Of Facial Micro-expressionsmentioning
confidence: 99%
“…Among appearance-based feature extraction methods, local binary pattern on three orthogonal planes (LBP-TOP) is widely applied in many works (Li et al, 2013; Guo et al, 2014; Le Ngo et al, 2014, 2015, 2016a,b; Yan et al, 2014a; Adegun and Vadapalli, 2016; Zheng et al, 2016; Wang et al, 2017). Most existing datasets (SMIC, CASME II, SAMM) have all reported the LBP-TOP as their baseline evaluation method.…”
Section: Recognition Of Facial Micro-expressionsmentioning
confidence: 99%
“…Moreover, in the person dependent experiments (fivefold cross validation), the average accuracy of 97.4% was achieved by 3D-CNNs plus transfer learning, which was more than twenty percent higher than person independent experiments (LOSO). In the results listed in Table 5, LBP-TOP + Nearest Neighbor [5] carried out person independent experiments and person dependent experiments, and the recognition accuracy of person dependent experiment was 65.8%, while the performance of person independent experiment was 53.7%.…”
Section: Smicmentioning
confidence: 99%
“…Method Accuracy LBP-SIP+RBF kernel [14] 64.02% OSW-LBP-TOP+RBF kernel [15] 57.71% LBP-TOP+Nearest Neighbor [17] 65.83% DLSTD+RPCA+SVM [16] 68.29% LBP-TOP+GSLSR 70.12% LBP-TOP+RPCA+GSLSR 67.07%…”
Section: Smic Databasementioning
confidence: 99%
“…As we did in the above experiment, we extracted LBP-TOP features and the leave-one-out cross validation was applied. For comparison purpose, four works are used as the baseline methods, i.e., in addition to the research work in [14], [15] and [16], we also choose the recent work of Guo et al in [17]. Besides, our method plus RPCA scheme proposed in [16] is included in the comparison as well.…”
Section: Smic Databasementioning
confidence: 99%