Automatic people segmentation with a template-driven graph cut

An image can be considered as a group of nodes as vertices and edges as links. Various techniques are formed based upon this assumption and energy minimization. Graph cut is one of the promising techniques for image segmentation. Boykov and Kolmogorov use mincut/ maxflow graph principal for image segmentation. We have discussed some other techniques such as grab cut which is user interactive and very effective technique. In last the current problems were also highlighted.

Procedings of the British Machine Vision Conference 2008, 2008

The graph cut based approach has become very popular for interactive segmentation of the object of interest from the background. One of the most important and yet largely unsolved issues in the graph cut segmentation framework is parameter selection. Parameters are usually fixed beforehand by the developer of the algorithm. There is no single setting of parameters, however, that will result in the best possible segmentation for any general image. Usually each image has its own optimal set of parameters. If segmentation of an image is not as desired under the current setting of parameters, the user can always perform more interaction until the desired results are achieved. However, significant interaction may be required if parameter settings are far from optimal. In this paper, we develop an algorithm for automatic parameter selection. We design a measure of segmentation quality based on different features of segmentation that are combined using AdaBoost. Then we run the graph cut segmentation algorithm for different parameter values and choose the segmentation of highest quality according to our learnt measure. We develop a new way to normalize feature weights for the AdaBoost based classifier which is particularly suitable for our framework. Experimental results show a success rate of 95.6% for parameter selection.

This paper presents a new algorithm for human segmentation in images. The human silhouette is estimated in positive windows that are already obtained with an existing efficient detection method. This accurate segmentation uses the data previously computed in the detection. First, a pre-segmentation step computes the likelihood of contour segments as being a part of a human silhouette. Then, a contour segment oriented graph is constructed from the shape continuity cue and the prior cue obtained by the pre-segmentation. Segmentation is so posed as the computation of the shortest-path cycle which corresponds to the human silhouette. Additionally, the process is achieved iteratively to eliminate irrelevant paths and to increase the segmentation performance. The approach is tested on a human image database and the segmentation performance is evaluated quantitatively.

Scientific Reports, 2012

We present a scale-invariant, template-based segmentation paradigm that sets up a graph and performs a graph cut to separate an object from the background. Typically graph-based schemes distribute the nodes of the graph uniformly and equidistantly on the image, and use a regularizer to bias the cut towards a particular shape. The strategy of uniform and equidistant nodes does not allow the cut to prefer more complex structures, especially when areas of the object are indistinguishable from the background. We propose a solution by introducing the concept of a ''template shape'' of the target object in which the nodes are sampled non-uniformly and non-equidistantly on the image. We evaluate it on 2D-images where the object's textures and backgrounds are similar, and large areas of the object have the same gray level appearance as the background. We also evaluate it in 3D on 60 brain tumor datasets for neurosurgical planning purposes.

Computer Vision Theory and Applications, 2010

Human detection and segmentation is a challenging task owing to variations in human pose and clothing. The union of Histograms of Oriented Gradients based descriptors and of a Support Vector Machine classifier is a classic and efficient method for human detection in the images. Conversely, as often in detection, accurate segmentation of these persons is not performed. Many applications however need it. This paper tackles the problem of giving rise to information that will guide the final segmentation step. It presents a method which uses the union mention above to relate to each contour segment a likelihood degree of being part of a human silhouette. Thus, data previously computed in detection are used in the pre-segmentation. A human silhouette database was ceated for learning. Figure 1: Input image (a), contour image computed with Canny algorithm (b), contour pixels gathered in contour segments (c) and likely segments (d).

This paper deals with automatically segmenting a person from challenging videos using a pose detector. A state of the art pose detector is used to detect the pose of a person from a frame in the video sequence. The pose is used to extract color and optical flow features to train a conditional random field to provide segmentation on multiple frames. Location from the pose is used to refine the results. No additional training data is required by the method. We also show how the pose results can be improved by our model.

Image segmentation plays a crucial role in effective understanding of digital images. Past few decades saw hundreds of research contributions in this field. However, the research on the existence of general purpose segmentation algorithm that suits for variety of applications is still very much active. Among the many approaches in performing image segmentation, graph based approach is gaining popularity primarily due to its ability in reflecting global image properties. This paper critically reviews existing important graph based segmentation methods. The review is done based on the classification of various segmentation algorithms within the framework of graph based approaches. The major four categorizations we have employed for the purpose of review are: graph cut based methods, interactive methods, minimum spanning tree based methods and pyramid based methods. This review not only reveals the pros in each method and category but also explores its limitations. In addition, the review highlights the need for creating a database for benchmarking intensity based algorithms, and the need for further research in graph based segmentation for automated real time applications.

2008 Ieee 16th Signal Processing Communication and Applications Conference, 2008

A graph theoretic color image segmentation algorithm is proposed, in which the popular normalized cuts image segmentation method is improved with modifications on its graph structure. The image is represented by a weighted undirected graph, whose nodes correspond to over-segmented regions, instead of pixels, that decreases the complexity of the overall algorithm. In addition, the link weights between the nodes are calculated through the intensity similarities of the neighboring regions. The irregular distribution of the nodes, as a result of such a modification, causes a bias towards combining regions with high number of links. This bias is removed by limiting the number of links for each node. Finally, segmentation is achieved by bipartitioning the graph recursively according to the minimization of the normalized cut measure. The simulation results indicate that the proposed segmentation scheme performs quite faster than the traditional normalized cut methods, as well as yielding better segmentation results due to its regionbased representation.

Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2012

We present a generic framework for object segmentation using depth maps based on Random Forest and Graph-cuts theory, and apply it to the segmentation of human limbs in depth maps. First, from a set of random depth features, Random Forest is used to infer a set of label probabilities for each data sample. This vector of probabilities is used as unary term in − swap Graph-cuts algorithm. Moreover, depth of spatio-temporal neighboring data points are used as boundary potentials. Results on a new multi-label human depth data set show high performance in terms of segmentation overlapping of the novel methodology compared to classical approaches.

Applications of Digital Image Processing XXXII, 2009

Most modern tracking techniques assume that the object comprises a large percentage of the image frame, however when the object is contained in a small number of pixels tracking via feature based methods is difficult, because they require a dense feature set which does not exist within small regions. As an alternative to dynamic boundary based methods, which require only a boundary between the object and the background, but often fail in busy enviroments, we propose using a novel graph cuts implemenation to obtain a more robust segmentation. The push-relabel method was chosen because of its lower time complexity. In addition the algorithm was expanded to the RGB color-space. This is done by a probabilistic combination of the RGB pixel values. This addition, by using all the information captured by the camera, allow objects with similar appearances and objects with large variances in color to be segmented. The final addition made to the the push-relabel algorithm is an min-cut approximation method which runs in O(n) time. We show that this formulation of the graph cut algorithm allows for a fast and accurate segmentation at 30 frames per second.