Motion Estimation in the 3-D Gabor Domain

Feng, Mingyue; Reed, T.R.

doi:10.1109/tip.2007.901812

Cited by 6 publications

(4 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Applying the 3D Gabor transform for feature extraction attracted more interests in recent years [35]. Feng et al [36] used 3D Gabor for motion estimation with adjustable spatiotemporal resolution, Zhen et al [37] applied the 3D Gabor library for MRI tagging sheet extraction and tracking, Kepenekci et al [38] analyzed motion using the 3D Gabor kernels. In this work, we proposed to apply the 3D Gabor library to the 3D facial expressions for human emotion recognition.…”

Section: D Gabor Feature Extractionmentioning

confidence: 99%

Human emotional state recognition using 3D facial expression features

Tie¹

2021

Preprint

View full text Add to dashboard Cite

In recent years there has been a growing interest in improving all aspects of the interaction between human and computers. Emotion recognition is a new research direction in human-computer interaction (HCI) which is based on affective computing that is expected to significantly improve the quality of HCI system and communications. Most existing works address this problem using 2D features, but they are sensitive to head pose, clutter, and variations in lighting conditions. In light of such problems, two 3D visual feature based approaches are presented in this dissertation. First, we present a recognition method based on the Gabor library for real 3D visual features extraction and an improved kernel canonical correlation analysis (IKCCA) algorithm for emotion classification. Second, to reduce the computation cost and provide a more general approach, we propose using a fiducial points' controlled 3D face model to recognize human emotion from video sequences. An Elastic body spine (EBS) technique is applied for deformation feature extraction and a discriminative Isomap (D-Isomap) based classification is used for the final decision. The most significant contributions of this work are detecting and tracking fiducial points automatically from video sequences to construct a generic 3D face model, and the introduction of EBS deformation features for emotion recognition. The experimental results show the robustness and effectiveness of the proposed methods.

show abstract

Section: D Gabor Feature Extractionmentioning

confidence: 99%

Human emotional state recognition using 3D facial expression features

Tie¹

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Furthermore, 3D Gabor filters have limited accuracy for motion estimation under severe cluttered backgrounds and heavy noise conditions [27,30]. Therefore, inspired by the properties of 3D spatiotemporal Gabor filters, we seek to design suitable spatiotemporal filters for IR small target tracking problems, which may have favorable efficiency and robustness.…”

Section: Spatiotemporal Gabor Filtersmentioning

confidence: 99%

“…Complex tracking needs more intricate modeling, which not only includes spatial responses but also contains temporal responses. In light of previous research [27][28][29][30], 3D Gabor filters are just essentially spatiotemporal filters, which combine information over space and time and have been effectively used for motion analysis. Motivated by 3D Gabor filters, we seek to extend the 2D DoG filter to a 3D domain for IR small target tracking.…”

Section: Introductionmentioning

confidence: 99%

Spatiotemporal difference-of-Gaussians filters for robust infrared small target tracking in various complex scenes

Wang

Chen

2015

Appl. Opt.

View full text Add to dashboard Cite

Tracking infrared (IR) small targets is a vital component of many computer vision applications, including IR precise guidance, early warning, and IR remote sensing. Various complicated scenes, however, present significant challenges to the tracking task. To solve this problem, we present a novel 3D spatiotemporal difference-of-Gaussians (DoG) filter-based algorithm for tracking small targets in IR videos of various complex scenes. First, biologically inspired 3D DoG filters are proposed for IR small target tracking, which are capable of accounting for spatial and temporal information. Then, based on such filters, an effective and robust tracker is constructed to track the small targets, which are spatiotemporally distinguishable from background clutter. Extensive experiments show that our approach tracks the small targets accurately and robustly in realistic IR videos of complex backgrounds that present unique difficulties to other approaches.

show abstract

“…Saisan [3] applied a dynamic texture model to the dynamic textures that assumes stationary dynamic textures well segmented in space and time. Feng [4] used 3D Gabor filter to extract the texture features.…”

Section: Introductionmentioning

confidence: 99%

Dynamic Textures Retrieval Using the Integrated Wavelet-Based Surfacelet Transform

Liu

Jiang³

2012

AMM

View full text Add to dashboard Cite

To reduce the redundancy, a new integrated wavelet-based surfacelet transform(WBST) is proposed. The integrated WBST is constructed by combining the integrated 3D non-redundant wavelet with nonsubsampled directional filter banks. For 3D non-redundant wavelet transform, there are seven high frequency sub-bands and one high frequency sub-bands. Therefore, we design the integrated scheme with three high frequency sub-bands. In order to get the texture-spatial features, 3D local binary pattern (3D LBP) is used to describe the dynamic texture feature of low frequency sub-band. The mean and standard deviation of coefficients in each sub bands can be used as the high frequency features. Experiments show that the proposed algorithm using the integrated WBST outperforms the 3D WT.

show abstract

Motion Estimation in the 3-D Gabor Domain

Cited by 6 publications

References 38 publications

Human emotional state recognition using 3D facial expression features

Human emotional state recognition using 3D facial expression features

Spatiotemporal difference-of-Gaussians filters for robust infrared small target tracking in various complex scenes

Dynamic Textures Retrieval Using the Integrated Wavelet-Based Surfacelet Transform

Contact Info

Product

Resources

About