Dynamic Textures

By Gianfranco Doretto, Alessandro Chiuso, Ying Nian Wu, Stefano Soatto. 2003

1,013 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Signal Processing

How researchers see this paper

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

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Signal Processing

Citations in the corpus

Learning Energy-Based Spatial-Temporal Generative ConvNets for Dynamic Patterns (2019, methodology): “[1], in terms of SSIM on 17 dynamic texture videos in Table 1.”
MoCoGAN: Decomposing Motion and Content for Video Generation (2017, methodology): “Due to limitations in computation, data, and modeling tools, early video generation works focused on generating dynamic texture patterns [35, 42, 9, 43].”
Learning deep dynamical models from image pixels (2014, methodology): “Doretto et al. (2003) implemented a subspace identification routine to model dynamical textures from highdimensional pixel data.”
Cooperative Training of Descriptor and Generator Networks (2016, methodology): “We conduct experiments on generating dynamic textures [69] with temporal stationarity.”
Alternating Back-Propagation for Generator Network (2016, positive): “(2) Treat {Z t } as the training data, learn A and Q as in (Doretto et al., 2003).”
Sequence of the Most Informative Joints (SMIJ): A new representation for human skeletal action recognition (2012, background): “[34] to identify the system parameters for the joint-angle time-series of a given action sequence.”
Spatiotemporal Saliency in Dynamic Scenes (2010, methodology): “We adopt dynamic texture models, due to their versatility in modeling complex moving patterns and ability to replicate natural scene dynamics [11], [31].”
Video-Based Facial Micro-Expression Analysis: A Survey of Datasets, Features and Algorithms (2021, background): “DT has been widely studied in the computer vision ﬁeld and is traditionally deﬁned as image sequences of moving objects/scenes that exhibit certain stationarity properties in time (see e.g. [39]).”
Machine Recognition of Human Activities: A Survey (2008, methodology): “Newer methods to learn the model parameters [86] have made learning much more efficient than in the case of HMMs.”
Dynamic texture recognition (2001, methodology): “To perform the learning process we used the closed-form solution proposed in [14].”
Synthesizing Dynamic Patterns by Spatial-Temporal Generative ConvNet (2016, methodology): “There are a wide variety of dynamic patterns in video sequences, including dynamic textures [2] or textured motions [24] that exhibit statistical stationarity or stochastic repetitiveness in the temporal dimension, and action patterns that are non-stationary in either spatial or temporal domain.”
Anomaly detection in crowded scenes (2010, methodology): “We then propose the use of representations based on dynamic textures (DTs) [10].”
Accurate 3D action recognition using learning on the Grassmann manifold (2015, methodology): “The dynamic captured by the ARMA [52, 53] model during an action sequence M can be represented as: where z 2 R is a hidden state vector, A 2 R is the transition matrix and C 2 R 3 J d is the measurement matrix. w and v are noise components modeled as normal with mean equal to zero and covariance mat”
Modeling Video Dynamics with Deep Dynencoder (2014, methodology): “LDS can model videos with smooth motions and has been applied to computer vision tasks such as DT synthesis [9], video segmentation [10][7] and video classification [18][5][17].”
Bio-inspired Dynamic 3D Discriminative Skeletal Features for Human Action Recognition (2013, methodology): “Given the time series {yt}t=1, the task of computing the system parameters, (x0, μ, A,C,B,R), is referred to as system identification and several optimal [21, 23] and sub-optimal, but very efficient [9], methods have been proposed in literature.”
Higher Order SVD Analysis for Dynamic Texture Synthesis (2008, methodology): “Recently, the parametric model proposed by Doretto et al. [1], [9] was shown to be a valid approach for analysis/synthesis of dynamic textures.”