Learn More
In this paper, we propose to learn a discriminative and share-able feature transformation filter bank to transform local image patches (represented as raw pixel values) into features for scene image classification. The learned filters are expected to: (1) encode common visual patterns of a flexible number of categories; (2) encode discriminative and(More)
—In the last few years, deep learning has lead to very good performance on a variety of problems, such as object recognition, speech recognition and natural language processing. Among different types of deep neural networks, convolutional neural networks have been most extensively studied. Due to the lack of training data and computing power in early days,(More)
Matching pedestrians across multiple camera views known as human re-identification (re-identification) is a challenging problem in visual surveillance. In the existing works concentrating on feature extraction , representations are formed locally and independent of other regions. We present a novel siamese Long Short-Term Memory (LSTM) architecture that can(More)
We adopt Convolutional Neural Networks (CNN) as our parametric model to learn discriminative features and classifiers for local patch classification. As visually similar pixels are indistinguishable from local context, we alleviate such ambiguity by introducing a global scene constraint. We estimate the global potential in a non-parametric framework.(More)
In image labeling, local representations for image units are usually generated from their surrounding image patches, thus long-range contextual information is not effectively encoded. In this paper, we introduce recurrent neural networks (RNNs) to address this issue. Specifically, directed acyclic graph RNNs (DAG-RNNs) are proposed to process DAG-structured(More)
Neurogenesis in the dentate gyrus (DG) declines severely by middle age, potentially because of age-related changes in the DG microenvironment. We hypothesize that providing fresh glial restricted progenitors (GRPs) or neural stem cells (NSCs) to the aging hippocampus via grafting enriches the DG microenvironment and thereby stimulates the production of new(More)
Acute seizure (AS) activity in old age has an increased predisposition for evolving into temporal lobe epilepsy (TLE). Furthermore, spontaneous seizures and cognitive dysfunction after AS activity are often intense in the aged population than in young adults. This could be due to an increased vulnerability of inhibitory interneurons in the aged hippocampus(More)
Maintenance of neurogenesis in adult hippocampus is important for functions such as mood and memory. As exposure to unpredictable chronic stress (UCS) results in decreased hippocampal neurogenesis, enhanced depressive- and anxiety-like behaviors, and memory dysfunction, it is believed that declined hippocampal neurogenesis mainly underlies the behavioral(More)
Greatly waned neurogenesis, diminished microvasculature, astrocyte hypertrophy and activated microglia are among the most conspicuous structural changes in the aged hippocampus. Because these alterations can contribute to age-related memory and mood impairments, strategies efficacious for mitigating these changes may preserve cognitive and mood function in(More)
In order to encode the class correlation and class specific information in image representation, we propose a new local feature learning approach named Deep Discriminative and Shareable Feature Learning (DDSFL). DDSFL aims to hierarchically learn feature transformation filter banks to transform raw pixel image patches to features. The learned filter banks(More)