The Shape Boltzmann Machine: A strong model of object shape

Eslami, S. M. Ali; Heess, Nicolas; Winn, John

doi:10.1109/cvpr.2012.6247702

Cited by 63 publications

(52 citation statements)

References 31 publications

(23 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…HiRF is also related to Shape Boltzmann Machines [10] used for object segmentation in images. They locally constrain the allowed extent of dependencies between their model variables so as to respect image segmentation.…”

Section: Related Workmentioning

confidence: 99%

“…FTCHiRF enforces local constraints on temporal dependencies of the hidden variables. Similar to Shape Boltzmann Machines [10], we partition the first hidden layer h 1 such that every partition has access only to a particular temporal segment of the video. While the partitions of h 1 cannot directly connect to all video segments, their long-range dependencies are captured through connections to the second hidden layer h 2 .…”

Section: Formulation Of Hirfmentioning

confidence: 99%

See 1 more Smart Citation

HiRF: Hierarchical Random Field for Collective Activity Recognition in Videos

Amer

Peng

Todorovic

2014

Computer Vision – ECCV 2014

100

View full text Add to dashboard Cite

Abstract. This paper addresses the problem of recognizing and localizing coherent activities of a group of people, called collective activities, in video. Related work has argued the benefits of capturing long-range and higher-order dependencies among video features for robust recognition. To this end, we formulate a new deep model, called Hierarchical Random Field (HiRF). HiRF models only hierarchical dependencies between model variables. This effectively amounts to modeling higher-order temporal dependencies of video features. We specify an efficient inference of HiRF that iterates in each step linear programming for estimating latent variables. Learning of HiRF parameters is specified within the max-margin framework. Our evaluation on the benchmark New Collective Activity and Collective Activity datasets, demonstrates that HiRF yields superior recognition and localization as compared to the state of the art.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Formulation Of Hirfmentioning

confidence: 99%

HiRF: Hierarchical Random Field for Collective Activity Recognition in Videos

Amer

Peng

Todorovic

2014

Computer Vision – ECCV 2014

100

View full text Add to dashboard Cite

show abstract

“…Since generative deep learning models like DBM have the specific property of generating realistic samples, they can generate samples which are different from shapes in our training set [14]. Since the size of the shape is fixed according to the known resolution, we just need to perform the shape alignment.…”

Section: Shape Alignmentmentioning

confidence: 99%

Aircraft Reconstruction in High-Resolution SAR Images Using Deep Shape Prior

Dou

Diao

Sun

et al. 2017

IJGI

View full text Add to dashboard Cite

Contour shapes are important features. An accurate contour shape of a target can provide important prior information for applications, such as target recognition, which can improve the accuracy of target interpretation. In this paper, a synthetic aperture radar (SAR) target reconstruction method is proposed, which can be used to reconstruct the target by using shape priors to perform an accurate extraction of the contour shape feature. The method is divided into two stages. In the deep shape prior extraction stage, a generative deep learning modelling method is used to obtain deep shape priors. In the reconstruction stage, a novel coarse-to-fine pose estimation method combined with an optimization algorithm is proposed, which integrates deep shape priors into the process of reconstruction. Specifically, to address the issue of object rotation, candidate poses are obtained using the coarse pose estimation method, which avoids an exhaustive search of each pose. In addition, an energy function composed of a scattering term and shape term to combine the fine pose estimation, is constructed and optimized via an iterative optimization algorithm to achieve the goal of object reconstruction. To the best of our knowledge, this is the first time shape priors have been used to extract shape features of SAR targets. Experiments are conducted on a data set acquired by TerraSAR-X images and the results demonstrate the high accuracy and robustness of the proposed method.

show abstract

“…When it comes to the use of deep learning for generation tasks, we can find various models, such as a deep Boltzmann machines (DBM) [8], [9], a denoising auto-encoder (DAE) [10], a shape Boltzmann machine (ShapeBM) [11], and a sum-product network (SPN) [12]. These models were mainly introduced to capture high-order abstractions for good representation of the observations, rather than for discriminative goal.…”

Section: Introductionmentioning

confidence: 99%

“…In the image generation experiments, we obtained much more realistic images generated from the DRM more than those from the other generative models. [11], and a sum-product network (SPN) [12]. These models were mainly introduced to capture high-order abstractions for good representation of the observations, rather than for discriminative goal.…”

mentioning

confidence: 99%

Deep Relational Model: A Joint Probabilistic Model with a Hierarchical Structure for Bidirectional Estimation of Image and Labels

Nakashika

2018

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYTwo different types of representations, such as an image and its manually-assigned corresponding labels, generally have complex and strong relationships to each other. In this paper, we represent such deep relationships between two different types of visible variables using an energy-based probabilistic model, called a deep relational model (DRM) to improve the prediction accuracies. A DRM stacks several layers from one visible layer on to another visible layer, sandwiching several hidden layers between them. As with restricted Boltzmann machines (RBMs) and deep Boltzmann machines (DBMs), all connections (weights) between two adjacent layers are undirected. During maximum likelihood (ML) -based training, the network attempts to capture the latent complex relationships between two visible variables with its deep architecture. Unlike deep neural networks (DNNs), 1) the DRM is a totally generative model and 2) allows us to generate one visible variables given the other, and 2) the parameters can be optimized in a probabilistic manner. The DRM can be also finetuned using DNNs, like deep belief nets (DBNs) or DBMs pre-training. This paper presents experiments conduced to evaluate the performance of a DRM in image recognition and generation tasks using the MNIST data set. In the image recognition experiments, we observed that the DRM outperformed DNNs even without fine-tuning. In the image generation experiments, we obtained much more realistic images generated from the DRM more than those from the other generative models. [11], and a sum-product network (SPN) [12]. These models were mainly introduced to capture high-order abstractions for good representation of the observations, rather than for discriminative goal. Once obtaining highlevel abstractions, we can, for instance, remove some noise on the observations, or restore missing parts in the observations.Most of the existing deep-learning approaches focus on extracting high-order abstractions from one variable. In this paper, we try to capture such high-order relationships between two different types of variables based on deep learning. For that, we introduced a probabilistic model called a deep relational model (DRM) [13]. A DRM is similar to an RBM and a DBM, each of which is a probabilistic model based on an energy function. The model sandwiches several hidden layers * * between two visible layers and defines a joint probability for the two visible variables. Every two adjacent layers are connected with undirected weights, which are estimated so as to maximize the likelihood of the two visible variables. Interestingly, since the DRM is a totally generative model, it allows us not only to apply it to recognition tasks, but to also generate samples of one variable from the other variable. For example, considering that we have two kinds of variables for a hand-written digit image and a one-hot vector of the labels, we can estimate the label by inferring mean-field posteriors given an image (classification task). On the other hand, by inferring posteriors g...

show abstract

The Shape Boltzmann Machine: A strong model of object shape

Cited by 63 publications

References 31 publications

HiRF: Hierarchical Random Field for Collective Activity Recognition in Videos

HiRF: Hierarchical Random Field for Collective Activity Recognition in Videos

Aircraft Reconstruction in High-Resolution SAR Images Using Deep Shape Prior

Deep Relational Model: A Joint Probabilistic Model with a Hierarchical Structure for Bidirectional Estimation of Image and Labels

Contact Info

Product

Resources

About