Semi-binary based video features for activity representation

Efficient and effective feature detection and representation is an important consideration when processing videos, and a large number of applications such as motion analysis, 3D scene understanding, tracking etc. depend on this. Amongst several feature description methods, local features are becoming increasingly popular for representing videos because of their simplicity and efficiency. While they achieve state-of-the-art performance with low computational complexity, their performance is still too limited for real world applications. Furthermore, rapid increases in the uptake of mobile devices has increased the demand for algorithms that can run with reduced memory and computational requirements. In this paper we propose a semi binary based feature detectordescriptor based on the BRISK detector, which can detect and represent videos with significantly reduced computational requirements, while achieving comparable performance to the state of the art spatio-temporal feature descriptors. First, the BRISK feature detector is applied on a frame by frame basis to detect interest points, then the detected key points are compared against consecutive frames for significant motion. Key points with significant motion are encoded with the BRISK descriptor in the spatial domain and Motion Boundary Histogram in the temporal domain. This descriptor is not only lightweight but also has lower memory requirements because of the binary nature of the BRISK descriptor, allowing the possibility of applications using hand held devices.We evaluate the combination of detectordescriptor performance in the context of action classification with a standard, popular bag-of-features with SVM framework. Experiments are carried out on two popular datasets with varying complexity and we demonstrate comparable performance with other descriptors with reduced computational complexity.

Text mining with semantic annotation : using enriched text representation for entity-oriented retrieval, semantic relation identification and text clustering

This project is a step forward in the study of text mining where enhanced text representation with semantic information plays a significant role. It develops effective methods of entity-oriented retrieval, semantic relation identification and text clustering utilizing semantically annotated data. These methods are based on enriched text representation generated by introducing semantic information extracted from Wikipedia into the input text data. The proposed methods are evaluated against several start-of-art benchmarking methods on real-life data-sets. In particular, this thesis improves the performance of entity-oriented retrieval, identifies different lexical forms for an entity relation and handles clustering documents with multiple feature spaces.

Supervised latent dirichlet allocation models for efficient activity representation

Local spatio-temporal features with a Bag-of-visual words model is a popular approach used in human action recognition. Bag-of-features methods suffer from several challenges such as extracting appropriate appearance and motion features from videos, converting extracted features appropriate for classification and designing a suitable classification framework. In this paper we address the problem of efficiently representing the extracted features for classification to improve the overall performance. We introduce two generative supervised topic models, maximum entropy discrimination LDA (MedLDA) and class- specific simplex LDA (css-LDA), to encode the raw features suitable for discriminative SVM based classification. Unsupervised LDA models disconnect topic discovery from the classification task, hence yield poor results compared to the baseline Bag-of-words framework. On the other hand supervised LDA techniques learn the topic structure by considering the class labels and improve the recognition accuracy significantly. MedLDA maximizes likelihood and within class margins using max-margin techniques and yields a sparse highly discriminative topic structure; while in css-LDA separate class specific topics are learned instead of common set of topics across the entire dataset. In our representation first topics are learned and then each video is represented as a topic proportion vector, i.e. it can be comparable to a histogram of topics. Finally SVM classification is done on the learned topic proportion vector. We demonstrate the efficiency of the above two representation techniques through the experiments carried out in two popular datasets. Experimental results demonstrate significantly improved performance compared to the baseline Bag-of-features framework which uses kmeans to construct histogram of words from the feature vectors.

Improving deep convultional neural networks with unsupervised feature learning

The latest generation of Deep Convolutional Neural Networks (DCNN) have dramatically advanced challenging computer vision tasks, especially in object detection and object classification, achieving state-of-the-art performance in several computer vision tasks including text recognition, sign recognition, face recognition and scene understanding. The depth of these supervised networks has enabled learning deeper and hierarchical representation of features. In parallel, unsupervised deep learning such as Convolutional Deep Belief Network (CDBN) has also achieved state-of-the-art in many computer vision tasks. However, there is very limited research on jointly exploiting the strength of these two approaches. In this paper, we investigate the learning capability of both methods. We compare the output of individual layers and show that many learnt filters and outputs of the corresponding level layer are almost similar for both approaches. Stacking the DCNN on top of unsupervised layers or replacing layers in the DCNN with the corresponding learnt layers in the CDBN can improve the recognition/classification accuracy and training computational expense. We demonstrate the validity of the proposal on ImageNet dataset.

Optimal feature extraction for bilingual OCR

Feature extraction in bilingual OCR is handicapped by the increase in the number of classes or characters to be handled. This is evident in the case of Indian languages whose alphabet set is large. It is expected that the complexity of the feature extraction process increases with the number of classes. Though the determination of the best set of features that could be used cannot be ascertained through any quantitative measures, the characteristics of the scripts can help decide on the feature extraction procedure. This paper describes a hierarchical feature extraction scheme for recognition of printed bilingual (Tamil and Roman) text. The scheme divides the combined alphabet set of both the scripts into subsets by the extraction of certain spatial and structural features. Three features viz geometric moments, DCT based features and Wavelet transform based features are extracted from the grouped symbols and a linear transformation is performed on them for the purpose of efficient representation in the feature space. The transformation is obtained by the maximization of certain criterion functions. Three techniques : Principal component analysis, maximization of Fisher's ratio and maximization of divergence measure have been employed to estimate the transformation matrix. It has been observed that the proposed hierarchical scheme allows for easier handling of the alphabets and there is an appreciable rise in the recognition accuracy as a result of the transformations.

HUMAN DETECTION USING SPARSE REPRESENTATION

The problem of human detection is challenging, more so, when faced with adverse conditions such as occlusion and background clutter. This paper addresses the problem of human detection by representing an extracted feature of an image using a sparse linear combination of chosen dictionary atoms. The detection along with the scale finding, is done by using the coefficients obtained from sparse representation. This is of particular interest as we address the problem of scale using a scale-embedded dictionary where the conventional methods detect the object by running the detection window at all scales.

Classification, representation, and automatic extraction of deformation features in sheet metal parts

This paper presents classification, representation and extraction of deformation features in sheet-metal parts. The thickness is constant for these shape features and hence these are also referred to as constant thickness features. The deformation feature is represented as a set of faces with a characteristic arrangement among the faces. Deformation of the base-sheet or forming of material creates Bends and Walls with respect to a base-sheet or a reference plane. These are referred to as Basic Deformation Features (BDFs). Compound deformation features having two or more BDFs are defined as characteristic combinations of Bends and Walls and represented as a graph called Basic Deformation Features Graph (BDFG). The graph, therefore, represents a compound deformation feature uniquely. The characteristic arrangement of the faces and type of bends belonging to the feature decide the type and nature of the deformation feature. Algorithms have been developed to extract and identify deformation features from a CAD model of sheet-metal parts. The proposed algorithm does not require folding and unfolding of the part as intermediate steps to recognize deformation features. Representations of typical features are illustrated and results of extracting these deformation features from typical sheet metal parts are presented and discussed. (C) 2013 Elsevier Ltd. All rights reserved.

Sparse Representation based Anomaly Detection using HOMV in H.264 Compressed Videos

In this paper, we have proposed an anomaly detection algorithm based on Histogram of Oriented Motion Vectors (HOMV) 1] in sparse representation framework. Usual behavior is learned at each location by sparsely representing the HOMVs over learnt normal feature bases obtained using an online dictionary learning algorithm. In the end, anomaly is detected based on the likelihood of the occurrence of sparse coefficients at that location. The proposed approach is found to be robust compared to existing methods as demonstrated in the experiments on UCSD Ped1 and UCSD Ped2 datasets.

A uniform phase representation for the harmonic model in speech synthesis applications

Feature-based vocoders, e.g., STRAIGHT, offer a way to manipulate the perceived characteristics of the speech signal in speech transformation and synthesis. For the harmonic model, which provide excellent perceived quality, features for the amplitude parameters already exist (e.g., Line Spectral Frequencies (LSF), Mel-Frequency Cepstral Coefficients (MFCC)). However, because of the wrapping of the phase parameters, phase features are more difficult to design. To randomize the phase of the harmonic model during synthesis, a voicing feature is commonly used, which distinguishes voiced and unvoiced segments. However, voice production allows smooth transitions between voiced/unvoiced states which makes voicing segmentation sometimes tricky to estimate. In this article, two-phase features are suggested to represent the phase of the harmonic model in a uniform way, without voicing decision. The synthesis quality of the resulting vocoder has been evaluated, using subjective listening tests, in the context of resynthesis, pitch scaling, and Hidden Markov Model (HMM)-based synthesis. The experiments show that the suggested signal model is comparable to STRAIGHT or even better in some scenarios. They also reveal some limitations of the harmonic framework itself in the case of high fundamental frequencies.

Feature Extraction Without Edge Detection

Information representation is a critical issue in machine vision. The representation strategy in the primitive stages of a vision system has enormous implications for the performance in subsequent stages. Existing feature extraction paradigms, like edge detection, provide sparse and unreliable representations of the image information. In this thesis, we propose a novel feature extraction paradigm. The features consist of salient, simple parts of regions bounded by zero-crossings. The features are dense, stable, and robust. The primary advantage of the features is that they have abstract geometric attributes pertaining to their size and shape. To demonstrate the utility of the feature extraction paradigm, we apply it to passive navigation. We argue that the paradigm is applicable to other early vision problems.

Multimodal lexicography: The representation of meaning in electronic dictionaries

One finding of user studies is that information on meaning tends to be what diction¬ary users want most from their dictionaries. This is consistent with the traditional image of the dictionary as a repository of meanings of words, and this is also borne out in definitions of the item DICTIONARY itself as given in dictionaries. While this popular view has not changed much, the growing role of electronic dictionaries can change the lexicographers' approach to meaning repre¬sentation. Traditionally, paper dictionaries have explained words with words, using either a defi¬nition or an equivalent, and occasionally a line-drawn picture. However, a prominent feature of the electronic medium is its multimodality, and this offers potential for the description of meaning. While it is much easier to include pictorial content, electronic dictionaries can also hold media objects which paper cannot carry, such as audio, animation or video. Publishers are drawn by the attraction of these new options, but are they always functionally useful for the dictionary users? In this article, the existing evidence is examined, and informed guesses are offered where evidence is not yet available.

Recognition of Human Action Using Moment-Based Feature

The performance of different classification approaches is evaluated using a view-based approach for motion representation. The view-based approach uses computer vision and image processing techniques to register and process the video sequence. Two motion representations called Motion Energy Images and Motion History Image are then constructed. These representations collapse the temporal component in a way that no explicit temporal analysis or sequence matching is needed. Statistical descriptions are then computed using moment-based features and dimensionality reduction techniques. For these tests, we used 7 Hu moments, which are invariant to scale and translation. Principal Components Analysis is used to reduce the dimensionality of this representation. The system is trained using different subjects performing a set of examples of every action to be recognized. Given these samples, K-nearest neighbor, Gaussian, and Gaussian mixture classifiers are used to recognize new actions. Experiments are conducted using instances of eight human actions (i.e., eight classes) performed by seven different subjects. Comparisons in the performance among these classifiers under different conditions are analyzed and reported. Our main goals are to test this dimensionality-reduced representation of actions, and more importantly to use this representation to compare the advantages of different classification approaches in this recognition task.

An Invariant Representation for Matching Trajectories across Uncalibrated Video Streams

We introduce a view-point invariant representation of moving object trajectories that can be used in video database applications. It is assumed that trajectories lie on a surface that can be locally approximated with a plane. Raw trajectory data is first locally approximated with a cubic spline via least squares fitting. For each sampled point of the obtained curve, a projective invariant feature is computed using a small number of points in its neighborhood. The resulting sequence of invariant features computed along the entire trajectory forms the view invariant descriptor of the trajectory itself. Time parametrization has been exploited to compute cross ratios without ambiguity due to point ordering. Similarity between descriptors of different trajectories is measured with a distance that takes into account the statistical properties of the cross ratio, and its symmetry with respect to the point at infinity. In experiments, an overall correct classification rate of about 95% has been obtained on a dataset of 58 trajectories of players in soccer video, and an overall correct classification rate of about 80% has been obtained on matching partial segments of trajectories collected from two overlapping views of outdoor scenes with moving people and cars.

Boundary, Brightness, and Depth Interactions During Preattentive Representation and Attentive Recognition of Figure and Ground

This article applies a recent theory of 3-D biological vision, called FACADE Theory, to explain several percepts which Kanizsa pioneered. These include 3-D pop-out of an occluding form in front of an occluded form, leading to completion and recognition of the occluded form; 3-D transparent and opaque percepts of Kanizsa squares, with and without Varin wedges; and interactions between percepts of illusory contours, brightness, and depth in response to 2-D Kanizsa images. These explanations clarify how a partially occluded object representation can be completed for purposes of object recognition, without the completed part of the representation necessarily being seen. The theory traces these percepts to neural mechanisms that compensate for measurement uncertainty and complementarity at individual cortical processing stages by using parallel and hierarchical interactions among several cortical processing stages. These interactions are modelled by a Boundary Contour System (BCS) that generates emergent boundary segmentations and a complementary Feature Contour System (FCS) that fills-in surface representations of brightness, color, and depth. The BCS and FCS interact reciprocally with an Object Recognition System (ORS) that binds BCS boundary and FCS surface representations into attentive object representations. The BCS models the parvocellular LGN→Interblob→Interstripe→V4 cortical processing stream, the FCS models the parvocellular LGN→Blob→Thin Stripe→V4 cortical processing stream, and the ORS models inferotemporal cortex.