Learning Intermediate-Level Representations of Form and Motion from Natural Movies

By Charles F. Cadieu, Barbara Olshausen. 2012

129 citations recorded by the source corpus.

Subfields: Cognitive Neuroscience, Computer Vision and Pattern Recognition

How researchers see this paper

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

In Cognitive Neuroscience

In Computer Vision and Pattern Recognition

Citations in the corpus

C3D: Generic Features for Video Analysis (2014, background): “Deep learning has also been applied to video feature learning in an unsuperivsed setting [5, 32].”
Unsupervised Feature Learning from Temporal Data (2015, background): “Several other studies combine the slowness prior with independence inducing priors [14, 4].”
Learning the Irreducible Representations of Commutative Lie Groups (2014, positive): “Motivated by a number of statistical phenomena observed in natural images, Cadieu & Olshausen (2012) describe a model that decomposes a signal into invariant amplitudes and covariant phase variables.”
Discovering Hidden Factors of Variation in Deep Networks (2014, methodology): “Previous approaches to separating factors of variation in data, such as content vs. style [3] or form vs. motion [4, 5, 6, 7, 8, 9], have relied upon a bilinear model architecture in which the units representing different factors are combined multiplicatively.”
Unsupervised Learning of Spatiotemporally Coherent Metrics (2014, background): “Several other studies combine the slowness prior with independence inducing priors [17, 5, 22].”
Learning to Relate Images (2011, background): “frames (see also [52] for a similar, probabilistic formulation”
Towards Nonlinear Disentanglement in Natural Data with Temporal Sparse Coding (2020, methodology): “To ﬁnd the best hyperparameters for the conditional prior regularization and the prior rate, we perform a random search over γ ∈ [1 , 16] and λ ∈ [1 , 10] and compute the recently proposed unsupervised disentanglement ranking (UDR) scores for unsupervised model selection [55].”
Unsupervised learning of depth and motion (2013, background): “Analogous models, also based on energy or crosscorrelation, have been proposed independently for motion encoding [3, 2, 12].”
Learning Image Representations Tied to Egomotion from Unlabeled Video (2017, background): “Typically, one seeks invariance to small transformations, e.g., the orientation binning and pooling operations in SIFT/HOG (Lowe 1999; Dalal and Triggs 2005; Tuytelaars andMikolajczyk 2008) andmodern CNNs both target invariance to local translations and rotations.”
Learning image representations equivariant to ego-motion (2015, background): “Temporal coherence has been explored for unsupervised feature learning with CNNs [20] [35] [7] [3] [17], with applications to dimensionality reduction [8], object recognition [20] [35], and metric learning [7].”
Unsupervised Learning of Visual Representations using Videos (2015, methodology): “The principle underlying ICA was first introduced by Herault and Jutten (1986) and stated more clearly by Comon (1994). The idea has since then been explored by a number of researchers (Bell and Sejnowski, 1995; Amari et al.”
A Review on Deep Learning Techniques for Video Prediction (2020, methodology): “The version we are reviewing is the one provided along with the work of Cadieu and Olshausen [127].”
Neural networks and neuroscience-inspired computer vision. (2014, background): “The analysis of time-varying signals for object perception has long been the subject of convergent interest across computational neuroscience and computer vision [76–86], but we argue that there is a good deal of room tomake these connections stronger going forward.”