Higher Order SVD Analysis for Dynamic Texture Synthesis

By Roberto Costantini, Luciano Sbaiz, Sabine Süsstrunk. 2008

128 citations recorded by the source corpus.

Subfields: Computational Mathematics, Computer Vision and Pattern Recognition, Statistical and Nonlinear Physics, Generative Models

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mathematics

1. Zoubin Ghahramani
2. Wei Chu
3. Dacheng Tao
4. Karl Meerbergen
5. Nuno Vasconcelos
6. Nick Vannieuwenhoven
7. Raf Vandebril
8. Antoni B. Chan

In Computer Vision and Pattern Recognition

1. Song-Chun Zhu
2. Ying Nian Wu
3. Dacheng Tao
4. Luc Van Gool
5. Ming-Hsuan Yang
6. Andrew Zisserman
7. Alan L. Yuille
8. Xiaogang Wang

In Statistical and Nonlinear Physics

1. Ying Nian Wu
2. Song-Chun Zhu
3. Jianwen Xie
4. Ruiqi Gao
5. Yang Lu
6. Tian Han
7. Dacheng Tao
8. José Carlos Príncipe

In Generative Models

1. Song-Chun Zhu
2. Ying Nian Wu
3. Jianwen Xie
4. Ruiqi Gao
5. Tian Han
6. Yang Lu
7. Erik Nijkamp
8. Mitch Hill

Citations in the corpus

• Learning Energy-Based Spatial-Temporal Generative ConvNets for Dynamic Patterns (2019, neutral): “We compare our method with some baseline methods for dynamic textures, such as HOSVD [69], FFT-LDS [70], and Doretto et al. [1], in terms of SSIM on 17 dynamic texture videos in Table 1.”
• Cooperative Training of Descriptor and Generator Networks (2016, background): “HOSVD [73]) in terms of PSNR and structural similarity measures (SSIM) on 6 dynamic texture videos.”
• Probabilistic Models for Incomplete Multi-dimensional Arrays (2009, background): “Costantini et al. (2008) showed that tensor decomposition on a dynamic texture yields a compact multipledimensional signal in the spatial, temporal and chromatic domains.”
• Spatio-Temporal Image Pattern Prediction Method Based on a Physical Model With Time-Varying Optical Flow (2013, background): “computer vision [1]–[4], image processing [5], [17], and other”
• Layered Dynamic Textures (2005, methodology): “To improve the DT’s ability for texture synthesis, Yuan et al. [15] model the hidden state variable with a closed-loop dynamic system that uses a feedback mechanism to correct errors with respect to a reference signal, while Costantini et al. [ 24 ] decompose videos as a multidimensional tensor usin”
• Kernel Learning for Dynamic Texture Synthesis (2016, methodology): “1) Sustainability Analysis: The visual quality comparison over two representative videos are shown in Figure 6.”
• A New Truncation Strategy for the Higher-Order Singular Value Decomposition (2012, methodology): “The HOSVD has been applied in numerous application domains [29], such as image processing [9, 32, 39, 61], pattern recognition [49, 50, 59, 60, 62], data mining [33, 34, 52, 53], signal processing [12, 19, 36, 37, 38, 45], psychometrics [54, 55, 56], chemometrics [5], and biomedicine [16, 40, 41].”
• Learning Generator Networks for Dynamic Patterns (2019, background): “A major class of models on dynamic textures focus on the state space auto-regressive model for synthesis [5, 9, 38, 4] and classification [2, 36].”