Search

Video Anomaly detection in videos refers to the identification of events that do not conform to expected behavior. However, almost all existing methods cast this problem as the minimization of reconstruction errors of training data including only normal events, which may lead to self-reconstruction and cannot guarantee a larger reconstruction error for an abnormal event. In this paper, we propose to formulate the video anomaly detection problem within a regime of video prediction. We advocate that not all video prediction networks are suitable for video anomaly detection. Then, we introduce two principles for the design of a video prediction network for video anomaly detection. Based on them, we elaborately design a video prediction network with appearance and motion constraints for video anomaly detection. Further, to promote the generalization of the prediction-based video anomaly detection for novel scenes, we carefully investigate the usage of a meta learning within our framework, where our model can be fast adapted to a new testing scene with only a few starting frames. Extensive experiments on both a toy dataset and three real datasets validate the effectiveness of our method in terms of robustness to the uncertainty in normal events and the sensitivity to abnormal events. [ABSTRACT FROM AUTHOR]

Lung metastasis, the most frequent metastatic pattern in hepatocellular carcinoma, is an important contributor to poor prognosis. However, the mechanisms responsible for lung metastasis in hepatocellular carcinoma remain unknown. Aiming to explore these mechanisms, weighted gene coexpression network analysis (WGCNA) was firstly used to find hub genes related to lung metastasis. Then, we obtained 67 genes related to lung metastasis in hepatocellular carcinoma which were mainly related to ribosomal pathways and functions, and a protein interaction network analysis identified that fibrillarin (FBL) might be an important hub gene. Furthermore, we found that FBL is highly expressed in hepatocellular carcinoma and that its high expression increases the rate of lung metastasis and indicates a poor prognosis. Knockdown of FBL could significantly reduce proliferation and stemness as well as inhibiting the migration and invasion of hepatocellular carcinoma cells. Moreover, we found that FBL might be involved in the regulation of MYC and E2F pathways in hepatocellular carcinoma. Finally, we demonstrated that the knockdown of FBL could suppress hepatocellular carcinoma cell growth in vivo. In conclusion, ribosome-biogenesis-related proteins, especially Fibrillarin, play important roles in lung metastasis from hepatocellular carcinoma. [ABSTRACT FROM AUTHOR]

This paper focus on analyzing graph connection of human joints for skeleton based video anomaly detection, which is more effective and efficient than those image-level reconstruction based or prediction based methods that may be affected by complex background. Specifically, we propose a spatial temporal graph convolutional networks based prediction network for skeleton based video anomaly detection. In other words, we build a normal graph describing graph connection of joints in normal data, where joints of abnormal events will be outliers of this graph. To our knowledge, this is the first work to apply graph convolutional networks on skeleton-based video anomaly detection. Experiments show that our proposed normal graph achieves the-state-of-art performance, compared to those image-level reconstruction-based or prediction-based methods, as well as RNN based methods upon joints. [ABSTRACT FROM AUTHOR]

Dictionary learning has played an important role in the success of sparse representation. Although several dictionary learning approaches have been developed for image classification, discriminative dictionary pair learning, i.e., jointly learning a synthesis dictionary and an analysis dictionary, is still in its infant stage. In this paper, we proposed a novel model of Fisher discrimination dictionary pair learning (FDDPL), in which Fisher discrimination information is embedded into analysis representation, analysis dictionary, and synthesis dictionary representation. With the proposed Fisher-like discrimination term, discrimination of both synthesis dictionary representation and analysis dictionary representation is introduced into the dictionary pair learning model. An iterative algorithm to efficiently solve the proposed FDDPL and a FDDPL based classifier are also presented in this paper. The experiments on face recognition, scene categorization, gender classification, and action recognition clearly show the advantage of the proposed FDDPL. [ABSTRACT FROM AUTHOR]

Anomaly detection in medical images refers to the identification of abnormal images with only normal images in the training set. Most existing methods solve this problem with a self-reconstruction framework, which tends to learn an identity mapping and reduces the sensitivity to anomalies. To mitigate this problem, in this paper, we propose a novel Proxy-bridged Image Reconstruction Network (ProxyAno) for anomaly detection in medical images. Specifically, we use an intermediate proxy to bridge the input image and the reconstructed image. We study different proxy types, and we find that the superpixel-image (SI) is the best one. We set all pixels’ intensities within each superpixel as their average intensity, and denote this image as SI. The proposed ProxyAno consists of two modules, a Proxy Extraction Module and an Image Reconstruction Module. In the Proxy Extraction Module, a memory is introduced to memorize the feature correspondence for normal image to its corresponding SI, while the memorized correspondence does not apply to the abnormal images, which leads to the information loss for abnormal image and facilitates the anomaly detection. In the Image Reconstruction Module, we map an SI to its reconstructed image. Further, we crop a patch from the image and paste it on the normal SI to mimic the anomalies, and enforce the network to reconstruct the normal image even with the pseudo abnormal SI. In this way, our network enlarges the reconstruction error for anomalies. Extensive experiments on brain MR images, retinal OCT images and retinal fundus images verify the effectiveness of our method for both image-level and pixel-level anomaly detection. [ABSTRACT FROM AUTHOR]

Dictionary learning has played an important role in the success of sparse representation. Although discriminative synthesis dictionary learning for sparse representation with a high-computational-complexity l 0 or l 1 norm constraint has been well studied for image classification, jointly and discriminatively learning an analysis dictionary and a synthesis dictionary is still in its infant stage. As a dual of synthesis dictionary, the recently developed analysis dictionary can provide a complementary view of data representation, which can have a much lower time complexity than sparse synthesis representation. Although several class-specific analysis-synthesis dictionary, which may have a big correlation between different classes' dictionaries, have been developed, how to learn a more compact and discriminative universal analysis-synthesis dictionary is still open. In this paper, to provide a more complete view of discriminative data representation, we propose a novel model of discriminative analysis-synthesis dictionary learning (DASDL), in which a linear classifier based on the coding coefficient is jointly learned with the dictionary pair, thus the performance of the classifier and the representational power of the dictionary pair being considered at the same time by the same optimization procedure. The size of the learned dictionaries can be very small since the analysis-synthesis dictionary is shared by all class data. An iterative algorithm to efficiently solve the proposed DASDL is presented in this paper. The experiments on face recognition, gender classification, action recognition and image classification clearly show the superiority of the proposed DASDL. [ABSTRACT FROM AUTHOR]

To simultaneously estimate the number of heads and locate heads with bounding boxes, we resort to detection-based crowd counting by leveraging RGB-D data and design a dual-path guided detection network (DPDNet). Specifically, to improve the performance of detection-based approaches for dense/tiny heads, we propose a density map guided detection module, which leverages density map to improve the head/non-head classification in detection network where the density implies the probability of a pixel being a head, and a depth-adaptive kernel that considers the variances in head sizes is also introduced to generate high-fidelity density map for more robust density map regression. In order to prevent dense heads from being filtered out during post-processing, we utilize such a density map for post-processing of head detection and propose a density map guided NMS strategy. Meanwhile, to improve the ability of detecting small heads, we also propose a depth-guided detection module to generate a dynamic dilated convolution to extract features of heads of different scales, and a depth-aware anchor is further designed for better initialization of anchor sizes in the detection framework. Then we use the bounding boxes whose sizes are generated with depth to train our DPDNet. Considering that existing RGB-D datasets are too small and not suitable for performance evaluation of data-driven based approaches, we collect two large-scale RGB-D crowd counting datasets, which comprise a synthetic dataset and a real-world dataset, respectively. Since the depth value at long-distance positions cannot be obtained in the real-world dataset, we further propose a depth completion method with meta learning, which fully utilizes the synthetic depth data to complete the depth value at long-distance positions. Extensive experiments on our proposed two RGB-D datasets and the MICC RGB-D counting dataset show that our method achieves the best performance for RGB-D crowd counting and localization. Further, our method can be easily extended to RGB image based crowd counting and achieves comparable or even better performance on the RGB datasets for both head counting and localization. [ABSTRACT FROM AUTHOR]

Background: The copper metabolism MURR1 domain (COMMD) protein family involved in tumor development and progression in several types of human cancer, but little is known about the function of COMMD proteins in hepatocellular carcinoma (HCC).Methods: The ONCOMINE and the UALCAN databases were used to evaluate the expression of COMMD1-10 in HCC and the association of this family with individual cancer stage and tumor grade. Kaplan-Meier (K-M) Plotter and Cox analysis hint the prognostic value of COMMDs. A network comprising 50 most similar genes and COMMD1-10 was constructed with the STRING database. Gene set enrichment analysis (GSEA) was performed using LinkedOmics database. The correlations between COMMD expression and the presence of immune infiltrating cells were also analyzed by the tumor immune estimation resource (TIMER) database. GSE14520 dataset and 80 HCC patients were used to validated the expression and survival value of COMMD3. Human HCC cell lines were also used for validating the function of COMMD3.Results: The expression of all COMMD family members showed higher expression in HCC tissues than that in normal tissues, and is associated with clinical cancer stage and pathological tumor grade. In HCC patients, the transcriptional levels of COMMD1/4 are positively correlated with overall survival (OS), while those of COMMD2/3/7/8/9 are negatively correlated with OS. Multivariate analysis indicated that a high level of COMMD3 mRNA is an independent prognostic factor for shorter OS in HCC patients. However, the subset of patients with grade 3 HCC, K-M survival curves revealed that high COMMD3/5/7/8/9 expression and low COMMD4/10 expression were associated with shorter OS. In addition, the expression of COMMD2/3/10 was associated with tumor-induced immune response activation and immune infiltration in HCC. The expression of COMMD3 from GSE14520 dataset and 80 patients are both higher in tumor than that in normal tissue, and a higher level of COMMD3 mRNA is associated with shorter OS. Knockdown of COMMD3 inhibits human HCC cell lines proliferation in vitro.Conclusions: Our study indicates that COMMD3 is an independent prognostic biomarker for the survival of HCC patients. COMMD3 supports the proliferation of HCC cells and contributes to the poor OS in HCC patients. [ABSTRACT FROM AUTHOR]

Video-based crowd counting can leverage the spatial-temporal information between neighboring frames, and thus this information would improve the robustness of crowd counting. Therefore, this solution is more practical than single image-based crowd counting in real applications. Since severe occlusions, translation, rotation, and scaling of persons will give rise to the change of density map of heads between neighboring frames, video-based crowd counting is a very challenging task. To alleviate these issues in video crowd counting, a Multi-Level Feature Fusion Based Locality-Constrained Spatial Transformer Network (MLSTN) is proposed, which consists of two components, namely density map regression module and Locality-Constrained Spatial Transformer (LST) module. Specifically, we first estimate the density map of each frame by utilizing the combination of the low-level, middle-level and high-level features of the Convolutional Neural Networks. This is because the low-level features may be more effective in the extraction of small head information, while the middle and high level features are more effective in the extraction of medium and large head information. Then to measure the relationship of the density maps between neighboring frames, the LST module is proposed, which estimates the density map of the next frame by concatenating several regression density maps. To facilitate the performance evaluation for video crowd counting, we have collected and labeled a large-scale video crowd counting dataset which includes 100 five-second-long sequences with 394,081 annotated heads from 13 different scenes. As far as we know, it is the largest video crowd counting dataset. Extensive experiments show the effectiveness of our proposed approach for crowd counting on our dataset and other video-based crowd counting datasets. All our dataset are released online. 1 1 https://github.com/sweetyy83/Lstn_fdst_dataset. [ABSTRACT FROM AUTHOR]

Compatibility property, as well as crystalline morphology, of NBR/PEO blends has been investigated by Fourier transform infrared spectroscopy (FTIR), differential scanning calorimetry (DSC), X-ray diffraction (XRD), and polarized optical microscopy (POM) thoroughly. There is no apparent shift of nitrile or ether groups in the FTIR spectra of NBR/PEO blends. Based on the calculations from glass transition temperature, the maximum volume fraction of PEO dissolved in NBR phase is about 6.41 % in blend with 5 wt% PEO content (PEO-5), indicating a weak intermolecular interaction in the NBR/PEO blends. From the characteristic absorption bands in the FTIR spectra, XRD and POM graphs, the crystallinity ratio of NBR/PEO blends decreases as the NBR content increases, which is further proved by DSC measurement that the crystallinity ratio and crystal melting temperature of pure PEO are 82.6 %, 69.9 °C, and that of PEO-5 are 16.9 %, 59.5 °C. This illuminates that the weak intermolecular interaction will affect the crystallinity ratio and crystal melting temperature of the NBR/PEO blends. [ABSTRACT FROM AUTHOR]

• The modification of slurry oil for de-oiled asphalt leads to a more reasonable chemical composition. • Radial SBS is efficient on improving high temperature performance of asphalt blends. • The performance of blends with 10% CR is comparable to that containing 5% SBS. • Microstructure and micromorphology of blends verified their mechanical response. The demand for high quality asphalt binder is continuously increasing in road construction and paving industry, most of which derives from vacuum distillation of petroleum. Meanwhile, deoiled asphalt (DOA) and slurry oil are the by-products of solvent deasphalting and fluid catalytic cracking (FCC) respectively. This paper developed high-quality road asphalt binder from DOA modified with FCC slurry oil and polymers. Different styrene-butadienestyrene (SBS) and crumb rubber (CR) were selected as polymer modifiers. The optimized blending ratio of DOA/slurry was determined. Moreover, the effects of various modifiers on rheological properties, low-temperature cracking resistance, and microstructure of the resulting asphalt were evaluated. The results indicated that the incorporation of FCC slurry oil results in a more reasonable chemical composition and the improved cracking resistance of the blends. For different structure of SBS, radial SBS is more efficient to enhance the high temperature performance of DOA/slurry blends. The DOA/slurry with 10% CR shows the approximate value in viscosity, complex modulus and |G*|/sinδ in comparison of sample with 5% linear SBS, demonstrating significant effectiveness of CR on improving the high temperature performance of DOA/slurry. Linear SBS is a better candidate improving low temperature performance of DOA/slurry. The FTIR analysis indicated that the double bonds of SBS take part in the chemical cross-linking reaction, whereas CR improves property of asphalt only by physical interactions. Regarding to morphology, linear SBS presents the homogeneous dispersion in the form of cycle particles, whereas radial SBS displays a rather irregular form of polymer phase. CR phase shows elongated features and forms a continuous phase when reaching 20% content, which significantly reinforces the mechanical property of the asphalt. [ABSTRACT FROM AUTHOR]

The utilization of nanomaterials in the field of binder materials for road paving has attracted researchers' attention in recent years. This study presented the performance properties of a binder modified with carbon nanotubes (CNT) and polyethylene (PE). The rheological properties, adhesion behavior, morphology, and storage stability of the modified asphalt were investigated. Experimental analysis indicated a positive effect of CNT/PE composites on the performance of the binder. The results indicate that the combined use of CNT and PE shows a significant enhancement on complex modulus, viscosity, and creep recovery of the binder at high temperatures and a great decrease in compliance, indicating great resistance to permanent deformation. Meanwhile, only using CNT to improve the high temperature performance of the binder is limited due to high shear mixing. CNT/PE modifiers enhance the cracking resistance at low temperatures and moisture damage resistance. The CNT/PE melt mixing composites endow asphalt with stronger cracking resistance, better resistance to moisture damage and workability. Asphalt with CNT/PE composites formed an even dispersion system. Notably, CNT bridges on the interface between PE phase and asphalt for the two modified asphalts, which reinforces the cohesion of interface. Asphalt with CNT/PE composites showed improved storage stability in comparison with PE modified asphalt. [ABSTRACT FROM AUTHOR]