Multi-class image segmentation using conditional random fields and global classification

Plath, Nils; Toussaint, Marc; Nakajima, Shinichi

doi:10.1145/1553374.1553479

Cited by 124 publications

(98 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, the neighborhood information can be integrated into the affinity learning procedure. As proven by many other papers [22], [23], [24], [25], [26], this information complements purely local information to a great advantage. We use the finest level of hierarchical image segmentation as the basic unit instead of pixels.…”

Section: Related Workmentioning

confidence: 71%

Affinity Learning with Diffusion on Tensor Product Graph

Yang

Prasad

Latecki

2013

IEEE Trans. Pattern Anal. Mach. Intell.

139

127

View full text Add to dashboard Cite

Abstract-In many applications, we are given a finite set of data points sampled from a data manifold and represented as a graph with edge weights determined by pairwise similarities of the samples. Often the pairwise similarities (which are also called affinities) are unreliable due to noise or due to intrinsic difficulties in estimating similarity values of the samples. As observed in several recent approaches, more reliable similarities can be obtained if the original similarities are diffused in the context of other data points, where the context of each point is a set of points most similar to it. Compared to the existing methods, our approach differs in two main aspects. First, instead of diffusing the similarity information on the original graph, we propose to utilize the tensor product graph (TPG) obtained by the tensor product of the original graph with itself. Since TPG takes into account higher order information, it is not a surprise that we obtain more reliable similarities. However, it comes at the price of higher order computational complexity and storage requirement. The key contribution of the proposed approach is that the information propagation on TPG can be computed with the same computational complexity and the same amount of storage as the propagation on the original graph. We prove that a graph diffusion process on TPG is equivalent to a novel iterative algorithm on the original graph, which is guaranteed to converge. After its convergence we obtain new edge weights that can be interpreted as new, learned affinities. We stress that the affinities are learned in an unsupervised setting. We illustrate the benefits of the proposed approach for data manifolds composed of shapes, images, and image patches on two very different tasks of image retrieval and image segmentation. With learned affinities, we achieve the bull's eye retrieval score of 99.99 percent on the MPEG-7 shape dataset, which is much higher than the state-of-the-art algorithms. When the data points are image patches, the NCut with the learned affinities not only significantly outperforms the NCut with the original affinities, but it also outperforms state-of-the-art image segmentation methods.

show abstract

Section: Related Workmentioning

confidence: 71%

Affinity Learning with Diffusion on Tensor Product Graph

Yang

Prasad

Latecki

2013

IEEE Trans. Pattern Anal. Mach. Intell.

139

127

View full text Add to dashboard Cite

show abstract

“…CRFs have been used for a variety of classification problems, including natural language processing (Lafferty et al, 2001;Culotta and McCallum, 2004), computer vision (Kumar and Hebert, 2003;Tappen et al, 2007;Levin and Weiss, 2009;Plath et al, 2009), human activity recognition (Sminchisescu et al, 2005;Liao et al, 2007;Vail et al, 2007) and bioinformatics (Fu et al, 2009). …”

Section: Applications Of Crfsmentioning

confidence: 99%

Improving the Supervised Learning of Activity Classifiers for Human Motion Data

Zhao

Wang

Sukthankar

2013

Human Behavior Recognition Technologies

View full text Add to dashboard Cite

The ability to accurately recognize human activities from motion data is an important stepping stone toward creating many types of intelligent user interfaces. Many supervised learning methods have been demonstrated for learning activity classifiers from data; however, these classifiers often fail due to noisy sensor data, lack of labeled training samples for rare actions and large individual differences in activity execution. In this chapter, we introduce two techniques for improving supervised learning of human activities from motion data: 1) an active learning framework to reduce the number of samples required to segment motion traces and 2) an intelligent feature selection technique that both improves classification performance and reduces training time. We demonstrate how these techniques can be used to improve the classification of human household activities, an area of particular research interest since it facilitates the development of elder-care assistance systems to monitor household occupants.

show abstract

“…The construction of multi-scale CRFs was previously demonstrated in [2] and [3]. In these experiments, each image was segmented independently at multiple scales.…”

Section: Previous Approaches To Multiscale Crfsmentioning

confidence: 99%

“…In addition, it could effectively incorporate evidence from other image classifiers to perform better localization on unlabeled images. [2] …”

Section: Previous Approaches To Multiscale Crfsmentioning

confidence: 99%

See 1 more Smart Citation

Multiscale Conditional Random Fields for Semi-supervised Labeling and Classification

Duvenaud

Marlin

Murphy

2011

2011 Canadian Conference on Computer and Robot Vision

View full text Add to dashboard Cite

Abstract-Motivated by the abundance of images labeled only by their captions, we construct tree-structured multiscale conditional random fields capable of performing semisupervised learning. We show that such caption-only data can in fact increase pixel-level accuracy at test time. In addition, we compare two kinds of tree: the standard one with pairwise potentials, and one based on noisy-or potentials, which better matches the semantics of the recursive partitioning used to create the tree.

show abstract

Multi-class image segmentation using conditional random fields and global classification

Cited by 124 publications

References 12 publications

Affinity Learning with Diffusion on Tensor Product Graph

Affinity Learning with Diffusion on Tensor Product Graph

Improving the Supervised Learning of Activity Classifiers for Human Motion Data

Multiscale Conditional Random Fields for Semi-supervised Labeling and Classification

Contact Info

Product

Resources

About