View-Aware Image Object Compositing and Synthesis from Multiple Sources

Image compositing is widely used to combine visual elements from separate source images into a single image. Although recent image compositing techniques are capable of achieving smooth blending of the visual elements from different sources, most of them implicitly assume the source images are taken in the same viewpoint. In this paper, we present an approach to compositing novel image objects from multiple source images which have different viewpoints. Our key idea is to construct 3D proxies for meaningful components of the source image objects, and use these 3D component proxies to warp and seamlessly merge components together in the same viewpoint. To realize this idea, we introduce a coordinate-frame based single-view camera calibration algorithm to handle general types of image objects, a structure-aware cuboid optimization algorithm to get the cuboid proxies for image object components with correct structure relationship, and finally a 3D-proxy transformation guided image warping algorithm to stitch object components. We further describe a novel application based on this compositing approach to automatically synthesize a large number of image objects from a set of exemplars. Experimental results show that our compositing approach can be applied to a variety of image objects, such as chairs, cups, lamps, and robots, and the synthesis application can create novel image objects with significant shape and style variations from a small set of exemplars.     

Transferring Case Knowledge To Adaptation Knowledge: An Approach for Case-Base Maintenance

In this article we propose a case-base maintenance methodology based on the idea of transferring knowledge between knowledge containers in a case-based reasoning (CBR) system. A machine-learning technique, fuzzy decision-tree induction, is used to transform the case knowledge to adaptation knowledge. By learning the more sophisticated fuzzy adaptation knowledge, many of the redundant cases can be removed. This approach is particularly useful when the case base consists of a large number of redundant cases and the retrieval efficiency becomes a real concern of the user. The method of maintaining a case base from scratch, as proposed in this article, consists of four steps. First, an approach to learning feature weights automatically is used to evaluate the importance of different features in a given case base. Second, clustering of cases is carried out to identify different concepts in the case base using the acquired feature-weights knowledge. Third, adaptation rules are mined for each concept using fuzzy decision trees. Fourth, a selection strategy based on the concepts of case coverage and reachability is used to select representative cases. In order to demonstrate the effectiveness of this approach as well as to examine the relationship between compactness and performance of a CBR system, experimental testing is carried out using the Traveling and the Rice Taste data sets. The results show that the testing case bases can be reduced by 36 and 39 percent, respectively, if we complement the remaining cases by the adaptation rules discovered using our approach. The overall accuracies of the two smaller case bases are 94 and 90 percent, respectively, of the originals.     

Computation of madalines' sensitivity to input and weight perturbations

The sensitivity of a neural network's output to its input and weight perturbations is an important measure for evaluating the network's performance. In this letter, we propose an approach to quantify the sensitivity of Madalines. The sensitivity is defined as the probability of output deviation due to input and weight perturbations with respect to overall input patterns. Based on the structural characteristics of Madalines, a bottom-up strategy is followed, along which the sensitivity of single neurons, that is, Adalines, is considered first and then the sensitivity of the entire Madaline network. By means of probability theory, an analytical formula is derived for the calculation of Adalines' sensitivity, and an algorithm is designed for the computation of Madalines' sensitivity. Computer simulations are run to verify the effectiveness of the formula and algorithm. The simulation results are in good agreement with the theoretical results.     

The synthesis of elastin, collagen, and glycosaminoglycans by high density primary cultures of neonatal rat aortic smooth muscle. An ultrastructural and biochemical study

Primary cultures of neonatal rat aortic smooth muscle cells inoculated at high densities (1 X 10(6) cells/25 cm2 Falcon flask) with adequate nutrient media and pH control grow rapidly and form multilayers of cells with typical "hill and valley" organization. After 10 days growth insoluble elastin formation could be visualized by phase contrast microscopy as small particles which grew rapidly to become larger irregular refractile aggregates and later coalesced to form larger aggregates and small fibres. With light and electronmicroscopy, elastin was the predominant matrix protein formed, with the "hill regions" of cultures containing abundant elastin aggregates and some collagen. In 2-week-old cultures differentiation could be observed within the cell multilayer. The older deeper cells contained more protein synthesis organelles and myofilaments and were in close association with large often coalescing elastin aggregates; compared to younger more superficial cells which contained more free polyribosomes less myofilaments, and were associated with fewer and small elastin aggregates. In older cultures this differentiation was not apparent; the cells contained many myofilaments, dense bodies, and lysosomes. Elastin aggregates and newly formed elastic fibres were abundant in the matrix. Quantitative analysis of insoluble elastin formation in the cell layer during the 4-week culture period indicated continuous biosynthesis and deposition which paralleled that of desmosine formation. Amino-acid analysis of a hot alkali insoluble residue (regarded as elastin) from 30-day-old cultures gave a profile identical with neonatal rat aortic elastin in vivo. Insoluble collagen formation in the cell layer tended to plateau after the log phase of growth was completed (10 days). Proteoglycans were found predominantly in the supernatant media. Glycosaminoglycan analysis revealed a profile of dermatan sulphate (32%), chondroitin 4-sulphate (43%), keratan and heparan sulphate (30%), with only a trace of hyaluronic acid. This study indicates that primary cultures of neonatal rat aortic smooth muscle cells remain differentiated in culture and have the unique capacity to continue to synthesize and deposit large amounts (mg) of insoluble elastin which aggregate and from elastic fibres in vitro.     

A separation theorem for single-source network coding

In this paper, we consider a point-to-point communication network of discrete memoryless channels. In the network, there are a source node and possibly more than one sink node. Information is generated at the source node and is multicast to each sink node. We allow a node to encode its received information before loading it onto an outgoing channel, where the channels are independent of each other. We also allow the nodes to pass along messages asynchronously. In this paper, we characterize the admissibility of single-source multi-sink communication networks. Our result can be regarded as a network generalization of Shannon's result that feedback does not increase the capacity of a discrete memoryless channels (DMCs), and it implies a separation theorem for network coding and channel coding in such a communication network.     

Optimal power control system of a research nuclear reactor

The power control system is a key control system for a nuclear reactor, which directly concerns the safe operation of a nuclear reactor. Much attention is paid to the power control system performance of nuclear reactor in engineering. The designers put a high value upon design of an optimal power control system. In this paper, a design method is applied to the design of power control system. According to the optimal control theory, an objective function, quadratic performance index with weight factors is proposed. Then, the objective function is transformed into frequency domain form by use of Paserval's theorem. In frequency domain, an optimal transfer function can be obtained at the lowest value of objective function. The system with optimal transfer function has an optimal performance. The transfer function of the power control system is derived from a typical research nuclear reactor. Using the state feedback theory, the transfer function is synthesized to the optimal transfer function. The simulative results with the optimal controller and with a conventional controller show that the performance of the optimal power control system is largely improved on dynamic characters. The method applied here not only can be used for research nuclear reactor but also can be easily extended to pressurized water reactor power plant and other fields.     

Designing Efficient and Robust Caching Algorithms for Streaming-on-Demand Services on the Internet

Content Delivery Networks (CDN) have been used on the Internet to cache media content so as to reduce the load on the original media server, network congestion, and latency. Due to the large size of media content compared to normal web objects, current caching algorithms used in the Internet are no longer suitable. This paper presents a high-performance prefetch system that accommodates user time-varying behavior. A hybrid caching technique, which combines prefetch and replacement algorithms, is also introduced. The robustness of the cache system against imperfect user request information is evaluated using three request noise models. Two prefetch performance indices are also presented to help content administrators in deciding when to update the user request profile for caching algorithms.     

Mobilization of peripheral blood stem cells in children using G-CSF after carboplatin containing myelosuppressive chemotherapy

PURPOSE: Peripheral blood stem cell (PBSC) apheresis provides an alternative to autologous marrow harvest as a source of hematologic stem cells for transplantation in children with solid tumors. PATIENTS AND METHODS: Eight children with metastatic or recurrent solid tumors underwent 27 apheresis procedures. Recovery from myelosuppressive chemotherapy occurred without continuous daily growth factor support prior to mobilization. Granulocyte colony stimulating factor (G-CSF) at 16 microgs/kg/day was used to increase stem cells in the peripheral circulation. CD 34 positive cells, mononuclear cells (MNC), and CFU-GM were measured in the apheresis products. Prior chemotherapy was examined as a clinical factor that affected PBSC yield. RESULTS: A significant correlation was found between CD 34+/kg and CFU-GM/kg of the products (r = 0.758, P < 0.001). Patients receiving cumulative doses of carboplatin over 1,600 mg/m2 produced adequate MNC (1 x 10(8)/kg) but yielded significantly less CD 34+ cells or CFU-GM than those patients receiving less carboplatin. Prior doses of etoposide and ifosfamide did not effect PBSC yield. CONCLUSIONS: The mobilization technique was well tolerated, and the products obtained produced trilineage engraftment in the patients that underwent peripheral blood stem cell transplantation. Peripheral blood stem cell apheresis in children can be optimized by selection of appropriate candidates and mobilization with G-CSF after an absence of hematopoietic growth factor support.     

Rapid and Superior Bacteria Killing of Carbon Quantum Dots/ZnO Decorated Injectable Folic Acid‐Conjugated PDA Hydrogel through Dual‐Light Triggered ROS and Membrane Permeability

One of the most difficult challenges in the biomedical field is bacterial infection, which causes tremendous harm to human health. In this work, an injectable hydrogel is synthesized through rapid assembly of dopamine (DA) and folic acid (FA) cross‐linked by transition metal ions (TMIs, i.e., Zn²⁺), which was named as DFT‐hydrogel. Both the two carboxyl groups in the FA molecule and catechol in polydopamine (PDA) easily chelates Zn²⁺ to form metal–ligand coordination, thereby allowing this injectable hydrogel to match the shapes of wounds. In addition, PDA in the hydrogel coated around carbon quantum dot‐decorated ZnO (C/ZnO) nanoparticles (NPs) to rapidly generate reactive oxygen species (ROS) and heat under illumination with 660 and 808 nm light, endows this hybrid hydrogel with great antibacterial efficacy against Staphylococcus aureus (S. aureus, typical Gram‐positive bacteria) and Escherichia coli (E. coli, typical Gram‐negative bacteria). The antibacterial efficacy of the prepared DFT‐C/ZnO‐hydrogel against S. aureus and E. coli under dual‐light irradiation is 99.9%. Importantly, the hydrogels release zinc ions over 12 days, resulting in a sustained antimicrobial effect and promoted fibroblast growth. Thus, this hybrid hydrogel exhibits great potential for the reconstruction of bacteria‐infected tissues, especially exposed wounds.