Locating people in video from semantic descriptions : a new database and approach

The location of previously unseen and unregistered individuals in complex camera networks from semantic descriptions is a time consuming and often inaccurate process carried out by human operators, or security staff on the ground. To promote the development and evaluation of automated semantic description based localisation systems, we present a new, publicly available, unconstrained 110 sequence database, collected from 6 stationary cameras. Each sequence contains detailed semantic information for a single search subject who appears in the clip (gender, age, height, build, hair and skin colour, clothing type, texture and colour), and between 21 and 290 frames for each clip are annotated with the target subject location (over 11,000 frames are annotated in total). A novel approach for localising a person given a semantic query is also proposed and demonstrated on this database. The proposed approach incorporates clothing colour and type (for clothing worn below the waist), as well as height and build to detect people. A method to assess the quality of candidate regions, as well as a symmetry driven approach to aid in modelling clothing on the lower half of the body, is proposed within this approach. An evaluation on the proposed dataset shows that a relative improvement in localisation accuracy of up to 21 is achieved over the baseline technique.

Medial temporal lobe roles in human path integration

Path integration is a process in which observers derive their location by integrating self-motion signals along their locomotion trajectory. Although the medial temporal lobe (MTL) is thought to take part in path integration, the scope of its role for path integration remains unclear. To address this issue, we administered a variety of tasks involving path integration and other related processes to a group of neurosurgical patients whose MTL was unilaterally resected as therapy for epilepsy. These patients were unimpaired relative to neurologically intact controls in many tasks that required integration of various kinds of sensory self-motion information. However, the same patients (especially those who had lesions in the right hemisphere) walked farther than the controls when attempting to walk without vision to a previewed target. Importantly, this task was unique in our test battery in that it allowed participants to form a mental representation of the target location and anticipate their upcoming walking trajectory before they began moving. Thus, these results put forth a new idea that the role of MTL structures for human path integration may stem from their participation in predicting the consequences of one's locomotor actions. The strengths of this new theoretical viewpoint are discussed.

The roles of adsorption in hydrate precipitation

It has been well established that organic compounds with adjacent hydroxyl groups in Bayer process liquor can inhibit gibbsite precipitation by acting as seed poisons. The degree of inhibition is a function of the number and stereochemistry of the hydroxyl groups. Seed poisons generally adsorb strongly onto hydrate surfaces, implying that surface coverage is the mechanism for yield inhibition. There are examples however of organics that strongly adsorb but do not lead to yield inhibition. There is a possibility that this apparent contradiction may be an artifact of differences in conditions between the adsorption and precipitation experiments. The present work investigates the adsorption and inhibition effects of a range of compounds under strictly similar conditions to clarify the role of adsorption on yield inhibition.

Semantic interference in object naming : an fMRI study of the postcue naming paradigm

It is well established that the time to name target objects can be influenced by the presence of categorically related versus unrelated distractor items. A variety of paradigms have been developed to determine the level at which this semantic interference effect occurs in the speech production system. In this study, we investigated one of these tasks, the postcue naming paradigm, for the first time with fMRI. Previous behavioural studies using this paradigm have produced conflicting interpretations of the processing level at which the semantic interference effect takes place, ranging from pre- to post-lexical. Here we used fMRI with a sparse, event-related design to adjudicate between these competing explanations. We replicated the behavioural postcue naming effect for categorically related target/distractor pairs, and observed a corresponding increase in neuronal activation in the right lingual and fusiform gyri-regions previously associated with visual object processing and colour-form integration. We interpret these findings as being consistent with an account that places the semantic interference effect in the postcue paradigm at a processing level involving integration of object attributes in short-term memory.

Semantic context and visual feature effects in object naming : an fMRI study using arterial spin labeling

Previous behavioral studies reported a robust effect of increased naming latencies when objects to be named were blocked within semantic category, compared to items blocked between category. This semantic context effect has been attributed to various mechanisms including inhibition or excitation of lexico-semantic representations and incremental learning of associations between semantic features and names, and is hypothesized to increase demands on verbal self-monitoring during speech production. Objects within categories also share many visual structural features, introducing a potential confound when interpreting the level at which the context effect might occur. Consistent with previous findings, we report a significant increase in response latencies when naming categorically related objects within blocks, an effect associated with increased perfusion fMRI signal bilaterally in the hippocampus and in the left middle to posterior superior temporal cortex. No perfusion changes were observed in the middle section of the left middle temporal cortex, a region associated with retrieval of lexical-semantic information in previous object naming studies. Although a manipulation of visual feature similarity did not influence naming latencies, we observed perfusion increases in the perirhinal cortex for naming objects with similar visual features that interacted with the semantic context in which objects were named. These results provide support for the view that the semantic context effect in object naming occurs due to an incremental learning mechanism, and involves increased demands on verbal self-monitoring.

Anterior temporal cortex and semantic memory : reconciling findings from neuropsychology and functional imaging

Studies of semantic impairment arising from brain disease suggest that the anterior temporal lobes are critical for semantic abilities in humans; yet activation of these regions is rarely reported in functional imaging studies of healthy controls performing semantic tasks. Here, we combined neuropsychological and PET functional imaging data to show that when healthy subjects identify concepts at a specific level, the regions activated correspond to the site of maximal atrophy in patients with relatively pure semantic impairment. The stimuli were color photographs of common animals or vehicles, and the task was category verification at specific (e.g., robin), intermediate (e.g., bird), or general (e.g., animal) levels. Specific, relative to general, categorization activated the antero-lateral temporal cortices bilaterally, despite matching of these experimental conditions for difficulty. Critically, in patients with atrophy in precisely these areas, the most pronounced deficit was in the retrieval of specific semantic information.

Exploring innovation driven value creation in B2B service firms : The roles of the manager, employees, and customers in value creation

This study adopts the premise that innovation capability underpins a service firm's value creation ability and that management style, employee behaviors and marketing underpin its innovation capability. This study examines the role of managers and employees in the creation and delivery of superior value to customers via the firm's innovation capability. To test this premise the current study examines the role of transformational leadership (TFL) as an aspect of the service firm's management style in creating and delivering value to customers through its services. This study adopts a multi-level study, collecting data from managers, employees and customers of service firms in a Southeast-Asian country, Cambodia. The results show that a service firm's innovation capability has a positive effect on the firm's value offering (VO), the VO has a positive relationship with customer perceived value-in use (PVI), and PVI has a positive relationship with firm performance. This study also finds moderating effects of TFL on the relationship between service innovation capability and VO, and of service marketing capability on the relationship between VO and PVI respectively.

The roles of JK-1 (FAM134B) expressions in colorectal cancer

The aims of the present study are to investigate the clinicopathological correlations of JK-1(FAM134B) expression and its relationship to carcinogenesis in a colorectal adenoma-adenocarcinoma model. JK-1(FAM134B) protein expression was studied in a colon cancer cell line by Western blot and immunocytochemistry. JK-1(FAM134B) expression profiles at mRNA and protein levels were investigated in cancer tissues from 236 patients with colorectal adenocarcinoma and 32 patients with colorectal adenoma using real-time polymerase chain reaction and immunohistochemistry. The findings were then correlated with the clinicopathological features of these tumours. JK-1(FAM134B) protein was demonstrated in the colon cancer cells by Western blot. The protein was located in the nuclei of the tumour cells at both cellular and tissue levels. In colorectal adenocarcinomas, lower levels of JK-1(FAM134B) protein expression were associated with younger age (p=0.032), larger tumour size (p=0.004), advanced cancer stages (p=0.016) and higher rates of cancer recurrence (p=0.04). Also, lower levels of JK-1(FAM134B) mRNA expression were associated with advanced cancer stages (p=0.02) and presence of lymphovascular invasion (p=0.014). Higher JK-1(FAM134B) mRNA and protein expression levels were identified in adenomas and non-neoplastic mucosae, compared to carcinomas (p=0.005). To conclude, JK-1(FAM134B) mRNA expression and JK1 (FAM134B) protein levels varied with the different stages of progression of colorectal tumours. The expression levels of the gene were associated with clinicopathological features in patients with colorectal adenocarcinoma suggesting that JK-1(FAM134B) gene has roles in controlling some steps in the development of the invasive phenotypes from colorectal adenoma to early staged as well as advanced staged colorectal adenocarcinomas.

miR-126 in human cancers : clinical roles and current perspectives

miR-126 has been implicated in the processes of inflammation and angiogenesis. Through these processes, miR-126 is implicated in cancer biology, but its role there has not been well reviewed. The aim of this review is to examine the molecular mechanisms and clinicopathological significance of miR-126 in human cancers. miR-126 was shown to have roles in cancers of the gastrointestinal tract, genital tracts, breast, thyroid, lung and some other cancers. Its expression was suppressed in most of the cancers studied. The molecular mechanisms that are known to cause aberrant expression of miR-126 include alterations in gene sequence, epigenetic modification and alteration of dicer abundance. miR-126 can inhibit progression of some cancers via negative control of proliferation, migration, invasion, and cell survival. In some instances, however, miR-126 supports cancer progression via promotion of blood vessel formation. Downregulation of miR-126 induces cancer cell proliferation, migration, and invasion via targeting specific oncogenes. Also, reduced levels of miR-126 are a significant predictor of poor survival of patients in many cancers. In addition, miR-126 can alter a multitude of cellular mechanisms in cancer pathogenesis via suppressing gene translation of numerous validated targets such as PI3K, KRAS, EGFL7, CRK, ADAM9, HOXA9, IRS-1, SOX-2, SLC7A5 and VEGF. To conclude, miR-126 is commonly down-regulated in cancer, most likely due to its ability to inhibit cancer cell growth, adhesion, migration, and invasion through suppressing a range of important gene targets. Understanding these mechanisms by which miR-126 is involved with cancer pathogenesis will be useful in the development of therapeutic targets for the management of patients with cancer.

Toward an object-based semantic memory for long-term operation of mobile service robots

Throughout a lifetime of operation, a mobile service robot needs to acquire, store and update its knowledge of a working environment. This includes the ability to identify and track objects in different places, as well as using this information for interaction with humans. This paper introduces a long-term updating mechanism, inspired by the modal model of human memory, to enable a mobile robot to maintain its knowledge of a changing environment. The memory model is integrated with a hybrid map that represents the global topology and local geometry of the environment, as well as the respective 3D location of objects. We aim to enable the robot to use this knowledge to help humans by suggesting the most likely locations of specific objects in its map. An experiment using omni-directional vision demonstrates the ability to track the movements of several objects in a dynamic environment over an extended period of time.

Self-efficacy moderation and mediation roles on BPSD and social support influences on subjective caregiver burden in Chinese spouse caregivers of dementia patients

Background: This study aims to explore moderation and mediation roles of caregiver self-efficacy between subjective caregiver burden and (a) behavioral and psychological symptoms (BPSD) of dementia; and (b) social support. Methods: A cross-sectional study with 137 spouse caregivers of dementia patients was conducted in Shanghai. We collected demographic information for the caregiver–patient dyads, as well as information associated with dementia-related impairments, caregiver social support, caregiver self-efficacy, and SF-36. Results: Multiple regression analysis showed that caregiver self-efficacy was a moderator both between BPSD and subjective caregiver burden, and social support and subjective caregiver burden. Results also showed a partial mediation effect of caregiver self-efficacy on the impact of BPSD on subjective caregiver burden, and a mediation effect of social support on subjective caregiver burden. Caregiver self-efficacy and subjective burden significantly influenced BPSD and social support. Conclusion: Caregiver self-efficacy played an important role in the paths by which the two factors influenced subjective burden. Enhancing caregiver self-efficacy for symptom management (particularly BPSD) can be an essential strategy for determining interventions to support dementia caregivers in China, and possibly in other countries.

Functions and therapeutic roles of exosomes in cancer

The role of exosomes in cancer development has become the focus of much research, due to the many emerging roles possessed by exosomes. These micro-vesicles that are ubiquitously released in to the extracellular milieu, have been found to regulate immune system function, particularly in tumorigenesis, as well as conditioning future metastatic sites for the attachment and growth of tumor tissue. Through an interaction with a range of host tissue, exosomes are able to generate a pro-tumor environment that is essential for carcinogenesis. Herein, we discuss the contents of exosomes and their contribution to tumorigenesis, as well as their role in chemotherapeutic resistance and the development of novel cancer treatments and the identification of cancer biomarkers.

Small and medium enterprises using software as a service : exploring the different roles of intermediaries

Software as a Service (SaaS) can provide significant benefits to small and medium enterprises (SMEs) due to advantages like ease of access, 7*24 availability, and utility pricing. However, underlying the SaaS delivery model is often the assumption that SMEs will directly interact with the SaaS vendor and use a self-service approach. In practice, we see the rise of SaaS intermediaries who can support SMEs with sourcing and leveraging SaaS. This paper reports on the roles of intermediaries and how they support SMEs with using SaaS. We conducted an empirical study of two SaaS intermediaries and analysed their business models, in particular their value propositions. We identified orientation (technology or customer) and alignment (operational or strategic) as themes for understanding their roles. The contributions of this paper include: (1) the identification and description of SaaS intermediaries for SMEs based on an empirical study and (2) understanding the different roles of SaaS intermediaries, in particular a more basic role based on technology orientation and operational alignment and a more value adding role based on customer orientation and strategic alignment. We propose that SaaS intermediaries can address SaaS adoption and implementation challenges of SMEs by playing a basic role and can also aim to support SMEs in creating business value with SaaS based solutions by playing an added value role.

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.