Deep learning from temporal coherence in video
Published 14 June 2009
Hossein Mobahi, Ronan Collobert, Jason Weston
Citations355
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
A learning method for deep architectures that takes advantage of sequential data, in particular from the temporal coherence that naturally exists in unlabeled video recordings, and is used to improve the performance on a supervised task of interest.
Abstract
This work proposes a learning method for deep architectures that takes advantage of sequential data, in particular from the temporal coherence that naturally exists in unlabeled video recordings. That is, two successive frames are likely to contain the same object or objects. This coherence is used as a supervisory signal over the unlabeled data, and is used to improve the performance on a supervised task of interest. We demonstrate the effectiveness of this method on some pose invariant object and face recognition tasks.
Keywords
Computer ScienceEngineering
Proceedings of the IEEEGradient-based learning applied to document recognition
58,219 Citations1998Yann LeCun, Léon Bottou +2 more
This paper reviews various methods applied to handwritten character recognition and compares them on a standard handwritten digit recognition task, and Convolutional neural networks are shown to outperform all other techniques.
The Nature of Statistical Learning Theory
39,279 Citations1995Vladimir Vapnik
ScienceNonlinear Dimensionality Reduction by Locally Linear Embedding
15,035 Citations2000Sam T. Roweis, Lawrence K. Saul
Locally linear embedding (LLE) is introduced, an unsupervised learning algorithm that computes low-dimensional, neighborhood-preserving embeddings of high-dimensional inputs that learns the global structure of nonlinear manifolds.
ScienceA Global Geometric Framework for Nonlinear Dimensionality Reduction
13,740 Citations2000Joshua B. Tenenbaum, Vin de Silva +1 more
An approach to solving dimensionality reduction problems that uses easily measured local metric information to learn the underlying global geometry of a data set and efficiently computes a globally optimal solution, and is guaranteed to converge asymptotically to the true structure.
The MIT Press eBooksSemi-Supervised Learning
4,308 Citations2006Olivier Chapelle, Schölkopf, B. +1 more
This first comprehensive overview of semi-supervised learning presents state-of-the-art algorithms, a taxonomy of the field, selected applications, benchmark experiments, and perspectives on ongoing and future research.
Learning a Similarity Metric Discriminatively, with Application to Face Verification
3,870 Citations2005Sumit Chopra, Raia Hadsell +1 more
The idea is to learn a function that maps input patterns into a target space such that the L/sub 1/ norm in the target space approximates the "semantic" distance in the input space.
IEEE Transactions on Neural NetworksFace recognition: a convolutional neural-network approach
3,090 Citations1997Sandra Lawrence, C. Lee Giles +2 more
A hybrid neural-network for human face recognition which compares favourably with other methods and analyzes the computational complexity and discusses how new classes could be added to the trained recognizer.
Parameterisation of a stochastic model for human face identification
2,586 Citations2002Ferdinando Samaria, Augustin Harter
This paper presents a set of experimental results in which various HMM parameterisations are analysed and shows that stochastic modelling can be used successfully to encode feature information.
Series in machine perception and artificial intelligenceSIGNATURE VERIFICATION USING A “SIAMESE” TIME DELAY NEURAL NETWORK
2,202 Citations1994Jane Bromley, JAMES W. BENTZ +6 more
International Journal of Pattern Recognition and Artificial IntelligenceSIGNATURE VERIFICATION USING A “SIAMESE” TIME DELAY NEURAL NETWORK
2,096 Citations1993Jane Bromley, JAMES W. BENTZ +6 more
An algorithm for verification of signatures written on a pen-input tablet based on a novel, artificial neural network called a "Siamese" neural network, which consists of two identical sub-networks joined at their outputs.
IEEE Transactions on Pattern Analysis and Machine Intelligence80 Million Tiny Images: A Large Data Set for Nonparametric Object and Scene Recognition
1,976 Citations2008Antonio Torralba, Rob Fergus +1 more
For certain classes that are particularly prevalent in the dataset, such as people, this work is able to demonstrate a recognition performance comparable to class-specific Viola-Jones style detectors.
Neural ComputationSlow Feature Analysis: Unsupervised Learning of Invariances
1,377 Citations2002Laurenz Wiskott, Terrence J. Sejnowski
Slow feature analysis (SFA) is a new method for learning invariant or slowly varying features from a vectorial input signal that is guaranteed to find the optimal solution within a family of functions directly and can learn to extract a large number of decor-related features, which are ordered by their degree of invariance.
Learning methods for generic object recognition with invariance to pose and lighting
1,332 Citations2004Yann LeCun, Fu Jie Huang +1 more
A real-time version of the system was implemented that can detect and classify objects in natural scenes at around 10 frames per second and proved impractical, while convolutional nets yielded 16/7% error.
Semi-Supervised Classification by Low Density Separation
710 Citations2005Olivier Chapelle, Alexander Zien +1 more
Three semi-supervised algorithms are proposed: deriving graph-based distances that emphazise low density regions between clusters, followed by training a standard SVM, and optimizing the Transductive SVM objective function by gradient descent.
Stochastic Gradient Learning in Neural Networks
565 Citations1991Léon Bottou
This chapter discusses connectionist learning algorithms, which consists of minimizing a cost of the form C(w) = E(J(z,w)dP(z) where dP is an unknown probability distribution that characterizes the problem to learn, and J, the loss function, defines the learning system itself.
NatureSelf-organizing neural network that discovers surfaces in random-dot stereograms
508 Citations1992Suzanna Becker, Geoffrey E. Hinton
The authors' simulations show that when the learning procedure is applied to adjacent patches of two-dimensional images, it allows a neural network that has no prior knowledge of the third dimension to discover depth in random dot stereograms of curved surfaces.
Deep learning via semi-supervised embedding
387 Citations2008Jason Weston, Frédéric Ratle +1 more
It is shown how nonlinear embedding algorithms popular for use with shallow semi-supervised learning techniques such as kernel methods can be applied to deep multilayer architectures, either as a regularizer at the output layer, or on each layer of the architecture.
On Manifold Regularization.
323 Citations2005Misha Belkin, Partha Niyogi +1 more
A semisupervised framework that incorporates labeled and unlabeled data in a generalpurpose learner and gives rise to a regularized form of spectral clustering with an out-of-sample extension is proposed.
Unsupervised learning : foundations of neural computation
273 Citations1999Geoffrey E. Hinton, Terrence J. Sejnowski
Lecture notes in computer scienceSynergistic Face Detection and Pose Estimation with Energy-Based Models
206 Citations2006Margarita Osadchy, Y. Le Cun +1 more
A novel method for real-time, simultaneous multi-view face detection and facial pose estimation that employs a convolutional network to map face images to points on a manifold, parametrized by pose, and non-face images to Points far from that manifold is described.
IEEE Transactions on Pattern Analysis and Machine IntelligenceReal-time focus range sensor
206 Citations1996Shree K. Nayar, Masahiro Watanabe +1 more
Neural ComputationLearning Optimized Features for Hierarchical Models of Invariant Object Recognition
168 Citations2003Heiko Wersing, Edgar Körner
This work proposes a feedforward model for recognition that shares components like weight sharing, pooling stages, and competitive nonlinearities with earlier approaches but focuses on new methods for learning optimal feature-detecting cells in intermediate stages of the hierarchical network.
Network Computation in Neural SystemsMutual information maximization: models of cortical self-organization
142 Citations1996Suzanna Becker
The Imax algorithms are based on the idea of maximizing the mutual information between the outputs of different network modules, and are capable of extracting higher-order features from data and are well suited to modelling intermediate-to-high-level perceptual processing stages.
IEEE Transactions on Pattern Analysis and Machine IntelligenceBuilding models of animals from video
102 Citations2006Deva Ramanan, David Forsyth +1 more
It is argued that tracking, object detection, and model building are all similar activities, and a fully automatic system that builds 2D articulated models known as pictorial structures from videos of animals is described.
View-based 3D object recognition with support vector machines
100 Citations2003Danny Roobaert, Marc M. Van Hulle
This work reports high correct classification of unseen views, especially considering that no domain knowledge is including into the proposed system, and suggests an active learning algorithm to reduce further the required number of training views.
Neural ComputationImplicit Learning in 3D Object Recognition: The Importance of Temporal Context
57 Citations1999Suzanna Becker
This model provides a tenable account of how people may perform 3D object recognition in a hierarchical, bottom-up fashion and a maximum likelihood cost function allows this scheme to be implemented in a biologically feasible, hierarchical neural circuit.
Action respecting embedding
53 Citations2005Michael Bowling, Ali Ghodsi +1 more
This paper examines the case where additional information is known about the data and shows how this variation can be solved with a semidefinite program, and evaluates the technique in a synthetic, robot-inspired domain, demonstrating qualitatively superior representations and quantitative improvements on a data prediction task.
Model generation for video-based object recognition
17 Citations2006Humera Noor, Shahid Hafeez Mirza +3 more
This paper first identifies the underlying topological structure of an image dataset containing different views of the objects and represents it as a neighborhood graph, which gives a smaller dataset leading to reduced space requirements and faster matching.
Neural Information Processing SystemsLearning Temporally Persistent Hierarchical Representations
12 Citations1996Suzanna Becker
The proposed biologically motivated model of cortical self-organization performed considerably better than supervised back-propagation at generalizing to novel views from a small number of training examples.