Quality based frame selection for video face recognition

Quality based frame selection is a crucial task in video face recognition, to both improve the recognition rate and to reduce the computational cost. In this paper we present a framework that uses a variety of cues (face symmetry, sharpness, contrast, closeness of mouth, brightness and openness of the eye) to select the highest quality facial images available in a video sequence for recognition. Normalized feature scores are fused using a neural network and frames with high quality scores are used in a Local Gabor Binary Pattern Histogram Sequence based face recognition system. Experiments on the Honda/UCSD database shows that the proposed method selects the best quality face images in the video sequence, resulting in improved recognition performance.

Visualisation of complex flows using texture-based techniques

Detailed representations of complex flow datasets are often difficult to generate using traditional vector visualisation techniques such as arrow plots and streamlines. This is particularly true when the flow regime changes in time. Texture-based techniques, which are based on the advection of dense textures, are novel techniques for visualising such flows. We review two popular texture based techniques and their application to flow datasets sourced from active research projects. The techniques investigated were Line integral convolution (LIC) [1], and Image based flow visualisation (IBFV) [18]. We evaluated these and report on their effectiveness from a visualisation perspective. We also report on their ease of implementation and computational overheads.

Measurement of structural changes to a food material during dehydration

Rates of dehydration/rehydration are important quality parameters for dried products. Theoretically, if there are no adverse effects on the integrity of the tissue structure, it should absorb water to the same moisture content of the initial product before drying.The purpose of this work is to semi-automate the process of detection of cell structure boundaries as a food is dehydrated and rehydrated. This will enable food materials researchers to quantify changes to material’s structure as these processes take place. Images of potato cells as they were dehydrated and rehydrated were taken using an electron microscope. Cell boundaries were detected using an image processing algorithm. Average cell area and perimeter at each stage of dehydration were calculated and plotted versus time. The results show that the algorithm can successfully identify cell boundaries.

Automatic particle picking of biological molecules imaged by electron microscopy

One of the next great challenges of cell biology is the determination of the enormous number of protein structures encoded in genomes. In recent years, advances in electron cryo-microscopy and high-resolution single particle analysis have developed to the point where they now provide a methodology for high resolution structure determination. Using this approach, images of randomly oriented single particles are aligned computationally to reconstruct 3-D structures of proteins and even whole viruses. One of the limiting factors in obtaining high-resolution reconstructions is obtaining a large enough representative dataset ($>100,000$ particles). Traditionally particles have been manually picked which is an extremely labour intensive process. The problem is made especially difficult by the low signal-to-noise ratio of the images. This paper describes the development of automatic particle picking software, which has been tested with both negatively stained and cryo-electron micrographs. This algorithm has been shown to be capable of selecting most of the particles, with few false positives. Further work will involve extending the software to detect differently shaped and oriented particles.

Realistic irrigation visualization in a surgical wound debridement simulator

Wound debridement refers to the removal of necrotic, devitalized, or contaminated tissue and/or foreign material to promote wound healing. Surgical debridement uses sharp instruments to cut dead tissue from a wound and it is the quickest and most efficient method of debridement. A wound debridement simulator [1,2] can ensure that a medical trainee is competent prior to performing a procedure on a genuine patient. Irrigation is performed at different stages of debridement in order to remove debris and reduce the bacteria count through rinsing the wound. This paper presents a novel approach for realistic irrigation visualization based on texture representations of debris. This approach applies image processing techniques to a series of images, which model the cleanliness of the wound. The active texture is generated and updated dynamically based on the irrigation state, location, and range. Presented results demonstrate that texture mapping and image processing techniques can provide effective and efficient solutions for irrigation visualization in the wound debridement simulator.

Learning personalized tag ontology from user tagging information

The cross-sections of the Social Web and the Semantic Web has put folksonomy in the spot light for its potential in overcoming knowledge acquisition bottleneck and providing insight for "wisdom of the crowds". Folksonomy which comes as the results of collaborative tagging activities has provided insight into user's understanding about Web resources which might be useful for searching and organizing purposes. However, collaborative tagging vocabulary poses some challenges since tags are freely chosen by users and may exhibit synonymy and polysemy problem. In order to overcome these challenges and boost the potential of folksonomy as emergence semantics we propose to consolidate the diverse vocabulary into a consolidated entities and concepts. We propose to extract a tag ontology by ontology learning process to represent the semantics of a tagging community. This paper presents a novel approach to learn the ontology based on the widely used lexical database WordNet. We present personalization strategies to disambiguate the semantics of tags by combining the opinion of WordNet lexicographers and users’ tagging behavior together. We provide empirical evaluations by using the semantic information contained in the ontology in a tag recommendation experiment. The results show that by using the semantic relationships on the ontology the accuracy of the tag recommender has been improved.

Personalization in tag ontology learning for recommendation making

Due to the explosive growth of the Web, the domain of Web personalization has gained great momentum both in the research and commercial areas. One of the most popular web personalization systems is recommender systems. In recommender systems choosing user information that can be used to profile users is very crucial for user profiling. In Web 2.0, one facility that can help users organize Web resources of their interest is user tagging systems. Exploring user tagging behavior provides a promising way for understanding users’ information needs since tags are given directly by users. However, free and relatively uncontrolled vocabulary makes the user self-defined tags lack of standardization and semantic ambiguity. Also, the relationships among tags need to be explored since there are rich relationships among tags which could provide valuable information for us to better understand users. In this paper, we propose a novel approach for learning tag ontology based on the widely used lexical database WordNet for capturing the semantics and the structural relationships of tags. We present personalization strategies to disambiguate the semantics of tags by combining the opinion of WordNet lexicographers and users’ tagging behavior together. To personalize further, clustering of users is performed to generate a more accurate ontology for a particular group of users. In order to evaluate the usefulness of the tag ontology, we use the tag ontology in a pilot tag recommendation experiment for improving the recommendation performance by exploiting the semantic information in the tag ontology. The initial result shows that the personalized information has improved the accuracy of the tag recommendation.

A survey of tagging techniques for music, speech and environmental sound

Sound tagging has been studied for years. Among all sound types, music, speech, and environmental sound are three hottest research areas. This survey aims to provide an overview about the state-of-the-art development in these areas.We discuss about the meaning of tagging in different sound areas at the beginning of the journey. Some examples of sound tagging applications are introduced in order to illustrate the significance of this research. Typical tagging techniques include manual, automatic, and semi-automatic approaches.After reviewing work in music, speech and environmental sound tagging, we compare them and state the research progress to date. Research gaps are identified for each research area and the common features and discriminations between three areas are discovered as well. Published datasets, tools used by researchers, and evaluation measures frequently applied in the analysis are listed. In the end, we summarise the worldwide distribution of countries dedicated to sound tagging research for years.

Timed and Probabilistic Automata for Automatic Animal Call Recognition

Automatic Call Recognition is vital for environmental monitoring. Patten recognition has been applied in automatic species recognition for years. However, few studies have applied formal syntactic methods to species call structure analysis. This paper introduces a novel method to adopt timed and probabilistic automata in automatic species recognition based upon acoustic components as the primitives. We demonstrate this through one kind of birds in Australia: Eastern Yellow Robin.

Deformable face ensemble alignment with robust grouped-L1 anchors

Many methods exist at the moment for deformable face fitting. A drawback to nearly all these approaches is that they are (i) noisy in terms of landmark positions, and (ii) the noise is biased across frames (i.e. the misalignment is toward common directions across all frames). In this paper we propose a grouped $\mathcal{L}1$-norm anchored method for simultaneously aligning an ensemble of deformable face images stemming from the same subject, given noisy heterogeneous landmark estimates. Impressive alignment performance improvement and refinement is obtained using very weak initialization as "anchors".

Efficient real-time face detection for high resolution surveillance applications

This paper presents an efficient face detection method suitable for real-time surveillance applications. Improved efficiency is achieved by constraining the search window of an AdaBoost face detector to pre-selected regions. Firstly, the proposed method takes a sparse grid of sample pixels from the image to reduce whole image scan time. A fusion of foreground segmentation and skin colour segmentation is then used to select candidate face regions. Finally, a classifier-based face detector is applied only to selected regions to verify the presence of a face (the Viola-Jones detector is used in this paper). The proposed system is evaluated using 640 x 480 pixels test images and compared with other relevant methods. Experimental results show that the proposed method reduces the detection time to 42 ms, where the Viola-Jones detector alone requires 565 ms (on a desktop processor). This improvement makes the face detector suitable for real-time applications. Furthermore, the proposed method requires 50% of the computation time of the best competing method, while reducing the false positive rate by 3.2% and maintaining the same hit rate.

Kernel analysis on Grassmann manifolds for action recognition

Modelling video sequences by subspaces has recently shown promise for recognising human actions. Subspaces are able to accommodate the effects of various image variations and can capture the dynamic properties of actions. Subspaces form a non-Euclidean and curved Riemannian manifold known as a Grassmann manifold. Inference on manifold spaces usually is achieved by embedding the manifolds in higher dimensional Euclidean spaces. In this paper, we instead propose to embed the Grassmann manifolds into reproducing kernel Hilbert spaces and then tackle the problem of discriminant analysis on such manifolds. To achieve efficient machinery, we propose graph-based local discriminant analysis that utilises within-class and between-class similarity graphs to characterise intra-class compactness and inter-class separability, respectively. Experiments on KTH, UCF Sports, and Ballet datasets show that the proposed approach obtains marked improvements in discrimination accuracy in comparison to several state-of-the-art methods, such as the kernel version of affine hull image-set distance, tensor canonical correlation analysis, spatial-temporal words and hierarchy of discriminative space-time neighbourhood features.

Improved foreground detection via block-based classifier cascade with probabilistic decision integration

Background subtraction is a fundamental low-level processing task in numerous computer vision applications. The vast majority of algorithms process images on a pixel-by-pixel basis, where an independent decision is made for each pixel. A general limitation of such processing is that rich contextual information is not taken into account. We propose a block-based method capable of dealing with noise, illumination variations, and dynamic backgrounds, while still obtaining smooth contours of foreground objects. Specifically, image sequences are analyzed on an overlapping block-by-block basis. A low-dimensional texture descriptor obtained from each block is passed through an adaptive classifier cascade, where each stage handles a distinct problem. A probabilistic foreground mask generation approach then exploits block overlaps to integrate interim block-level decisions into final pixel-level foreground segmentation. Unlike many pixel-based methods, ad-hoc postprocessing of foreground masks is not required. Experiments on the difficult Wallflower and I2R datasets show that the proposed approach obtains on average better results (both qualitatively and quantitatively) than several prominent methods. We furthermore propose the use of tracking performance as an unbiased approach for assessing the practical usefulness of foreground segmentation methods, and show that the proposed approach leads to considerable improvements in tracking accuracy on the CAVIAR dataset.

Speaker linking using complete-linkage clustering

Speaker diarization determines instances of the same speaker within a recording. Extending this task to a collection of recordings for linking together segments spoken by a unique speaker requires speaker linking. In this paper we propose a speaker linking system using linkage clustering and state-of-the-art speaker recognition techniques. We evaluate our approach against two baseline linking systems using agglomerative cluster merging (AC) and agglomerative clustering with model retraining (ACR). We demonstrate that our linking method, using complete-linkage clustering, provides a relative improvement of 20% and 29% in attribution error rate (AER), over the AC and ACR systems, respectively.

Speaker attribution of multiple telephone conversations using a complete-linkage clustering approach

In this paper we propose and evaluate a speaker attribution system using a complete-linkage clustering method. Speaker attribution refers to the annotation of a collection of spoken audio based on speaker identities. This can be achieved using diarization and speaker linking. The main challenge associated with attribution is achieving computational efficiency when dealing with large audio archives. Traditional agglomerative clustering methods with model merging and retraining are not feasible for this purpose. This has motivated the use of linkage clustering methods without retraining. We first propose a diarization system using complete-linkage clustering and show that it outperforms traditional agglomerative and single-linkage clustering based diarization systems with a relative improvement of 40% and 68%, respectively. We then propose a complete-linkage speaker linking system to achieve attribution and demonstrate a 26% relative improvement in attribution error rate (AER) over the single-linkage speaker linking approach.