Search

Implicit neural representation has emerged as a powerful method for reconstructing 3D scenes from 2D images. Given a set of camera poses and associated images, the models can be trained to synthesize novel, unseen views. In order to expand the use cases for implicit neural representations, we need to incorporate camera pose estimation capabilities as part of the representation learning, as this is necessary for reconstructing scenes from real-world video sequences where cameras are generally not being tracked. Existing approaches like COLMAP and, most recently, bundle-adjusting neural radiance field methods often suffer from lengthy processing times. These delays ranging from hours to days, arise from laborious feature matching, hardware limitations, dense point sampling, and long training times required by a multi-layer perceptron structure with a large number of parameters. To address these challenges, we propose a framework called bundle-adjusting accelerated neural graphics primitives (BAA-NGP). Our approach leverages accelerated sampling and hash encoding to expedite both pose refinement/estimation and 3D scene reconstruction. Experimental results demonstrate that our method achieves a more than 10 to 20 $\times$ speed improvement in novel view synthesis compared to other bundle-adjusting neural radiance field methods without sacrificing the quality of pose estimation.

To investigate the effect and mechanism of pinolenic acid (PNA) on H2O2‐induced oxidative stress injury in HepG2 cells. Methods: PNA was used to regulate oxidative stress injury of HepG2 cells induced by H2O2. Quantification of cell survival rate, accumulation of intracellular reactive oxygen species (ROS), and expression levels of anti‐oxidation‐related genes were determined using MTT, fluorescent probe technology (DCFH‐DA), and real‐time quantitative reverse transcription polymerase chain technology (qRT‐PCR) method, respectively. Meanwhile, the activity of intracellular antioxidant enzymes was determined by biochemical methods. The results showed that PNA improved the survival rate of HepG2 cells induced by H2O2 (29.59%, high‐dose group), reduced the accumulation of intracellular ROS (65.52%, high‐dose group), and reduced the level of intracellular malondialdehyde (MDA; 65.52%, high‐dose group). All these results were dose‐dependent, which indicated that PNA can improve oxidative stress damage of cells. Furthermore, the mechanism of PNA regulating oxidative stress was investigated from the gene level. Results showed that under supplementation of PNA, the activity of superoxide dismutase (SOD), catalase (CAT), and glutathione peroxidase (GSH‐Px) had been improved (39.74%, 17.58%, and 23.83%, high‐dose group). Further studies on gene expression which controls the activity of antioxidant enzymes showed that under the regulation of PNA, the expression level of Keap1 gene was decreased, while Nrf2 gene was increased. The expression levels of HO‐1 and NQO1 in the downstream of Nrf2 were increased. Results indicated that under the regulation of PNA, Nrf2 was separated from Keap1, entered the nucleus, bound to ARE, and up‐regulated the expression levels of HO‐1 and NQO1 genes. Conclusion: PNA has a conspicuous improvement effect on oxidative stress damage induced by H2O2 in HepG2 cells. We also found the antioxidant mechanisms of PNA where it protected cells from oxidative stress damage by causing nuclear translocation of Nrf2 gene and up‐regulated the expression levels of antioxidant enzymes in the downstream. This shows that PNA prevented oxidative stress by mediating the Keap1/Nrf2 transcriptional pathway and down‐regulating enzyme activities.

Additional file 1: Table S1. Size of Nanoparticles. Figure S1. The SEM images of FeCOF. Figure S2. The SEM images of FeCOF@CaCO3. Figure S3. (a) XPS spectra of FeCOF@CaCO3. (b) XPS high-resolution scans of Ca 2p. Figure S4. The PXRD of FeCOF and FeCOF@CaCO3. Figure S5. (a) N2 adsorption and desorption isotherms at 77 K of FeCOF and FeCOF@CaCO3. (b)Pore size distribution plots of FeCOF and FeCOF@CaCO3. Figure S6. The Hydrodynamic diameters of FeCOF, FeCOF@CaCO3 and FCCF. Figure S7. The zate potential of FeCOF, FeCOF@CaCO3 and FCCF. Figure S8. (a) The FCCF dispersed in water, PBS and cell culture medium (containing 10% serum) before and after 7 days incubation. (b) The TEM image of FCCF stored in cell culture medium (containing 10% serum) for a week. (c) The DLS was measured on day 0, 5 and 7 when FCCF was dispersed in water, PBS and cell culture medium (containing 10% serum). Figure S9. (a) SEM images of FCCF at pH 7.4, 6.5 and 5.5, related to Figure 2a. Data are presented as mean ± SD (n = 3). Figure S10. (a) SEM images of FeCOF at pH 7.4, 6.5 and 5.5. Data are presented as mean ± SD (n = 3). Figure S11. Oxygen release behaviour of OFCCF in solution. Figure S12. CLSM microimages of intracellular Ca2+ content in 4T1 cells. Figure S13. CLSM microimages of mitochondrial Ca2+ content in 4T1 cells. Figure S15. Quantitative analysis of the intracellular O2 generation of 4T1 cells after various treatments based on the confocal images shown in Figure 3b by using imageJ, related to Figure 3b. Figure S16. Quantitative analysis of HIF-1α protein expression, as the ratio of protein to β-actin from Western Blot results. Related to Figure 3c. P values were calculated by one-way analysis (*p

Performing single image holistic understanding and 3D reconstruction is a central task in computer vision. This paper presents an integrated system that performs dense scene labeling, object detection, instance segmentation, depth estimation, 3D shape reconstruction, and 3D layout estimation for indoor and outdoor scenes from a single RGB image. We name our system panoptic 3D parsing (Panoptic3D) in which panoptic segmentation ("stuff" segmentation and "things" detection/segmentation) with 3D reconstruction is performed. We design a stage-wise system, Panoptic3D (stage-wise), where a complete set of annotations is absent. Additionally, we present an end-to-end pipeline, Panoptic3D (end-to-end), trained on a synthetic dataset with a full set of annotations. We show results on both indoor (3D-FRONT) and outdoor (COCO and Cityscapes) scenes. Our proposed panoptic 3D parsing framework points to a promising direction in computer vision. Panoptic3D can be applied to a variety of applications, including autonomous driving, mapping, robotics, design, computer graphics, robotics, human-computer interaction, and augmented reality.

Additional file 1. Table S1 Search strategies of databases.

Aluminum industry is a typical energy-intensive and emission-intensive industry. Henan's aluminum output occupies the first for many years until 2013 in the whole country. We analyzed 18 applicable to aluminum smelting process and 8 energy-efficiency technologies to alumina production process. The Conservation Supply Curve (CSC) is used in this paper. It is an analytical tool which selects the economically feasible technologies. Three scenarios are simulated. Under the BAU, S1 and S2 scenario, the energy consumption of the aluminum industry will decrease by 19%, 25%, and 29% compared to 2014 level respectively. The emission mitigation of GHG in S1 and S2 scenario are 3.2 Mt CO2e and 5.4 Mt CO2e, compared to BAU scenario in 2030. In addition, sensitivity analysis is conducted. Finally, some policy implications are proposed.

An approach to determine energy consumption control target allocation based on analytic hierarchy process, Ward's clustering algorithm, and differential adjustment method is proposed in this study. A case study of the allocation of control targets in Henan by 2020 is then conducted via the proposed method. The results show that: The per capita added value of the secondary industry is the primary factor for the increasing carbon emissions in provinces. Regions with high cardinality of energy consumption have to shoulder the largest reduction, whereas regions with low energy intensity and poor economic conditions met the minimum requirements for energy consumption in 2020.

We tackle the problem of point cloud recognition.Unlike previous approaches where a point cloud is either converted into a volume/image or represented independently in a permutation-invariant set, we develop a new representation by adopting the concept of shape context as the building block in our network design. The resulting model, called ShapeContextNet, consists of a hierarchy with modules not relying on a fixed grid while still enjoying properties similar to those in convolutional neural networks --- being able to capture and propagate the object part information.In addition, we find inspiration from self-attention based models to include a simple yet effective contextual modeling mechanism --- making the contextual region selection, the feature aggregation, and the feature transformation process fully automatic. ShapeContextNet is an end-to-end model that can be applied to the general point cloud classification. We observe competitive results on a number of benchmark datasets.

At present, on the analysis of features of warehouse system of small and medium mechatronic enterprise at home, a integrated architecture of warehouse management system is proposed to improve the performance of those enterprises. It has three layers: management layer, control layer and execution layer. Information sub-system, dispatching sub-system, simulation sub-system, hand-held terminal sub-system and warehouse execution sub-system are constructed on this architecture. The key technologies, including multi-layer architecture based on B/S, data exchange technology based on XML, hand-held terminal and task-role-based access control model, are discussed in detail. Finally, an interface of the integrated system based on the architecture and key technologies is given and the real application is shown that it's good for economical and efficient improvement of small and medium mechatronic enterprise at home.

On analysis of several existed task-role-based access control models, a novel task-role-based access control model is proposed to provide a safe and easy way to access data in the workflow system. The definition of user, role, task and privilege, constraint rule, authorization strategy and their relationship are described formally. The privilege is defined as a series of operations on object in the access control model. The constraints of user and role are defined by the constraints of task. In this way, the relationship of user, role and privilege is described clearly and it's convenient for system maintenance and safety management. The implementation of the model in practical shows that it could provide a more clear way to describe the access privilege of one user in a role to execute one task in the system. It could also describe the achievement of privileges in the workflow system.

The conductive mechanism is discussed in this paper. The calculation formula of Q and square resistance value is given to express the increase of RFID antennapsilas work efficiency. The influence factor of printing RFID antenna using conductive ink is given according to work feature and requirement of RFID antenna and lots of experiment data. These factors are conductive performance of metal granule in conductive ink, metal granule content in ink, solvent volatilization and design figure of antenna.

AGV as one part of automated warehouse system, its role is goods transportation. The dispatching problem of AGV sub-system in automated warehouse system is analyzed. The dispatching mechanism of AGV was discussed in detail. A math model was built on analysis of the optimization problem. A method with genetic algorithm was proposed to solve the optimization problem. Coding, selection and mutation was given in detail. The simulation result shows that the approach is feasible and provides a good practical way for solving optimization problem in automated warehouse system.

On the basis of discussion of RFID system work principle, the method of calculation of minimum action magnetic intensity and its impact factor are proposed. To the tag read-write angle, the relationship between tag read-write range and its loss detection is analyzed. The value of detection rate on conveyor belt is tested by experiment. Two methods of improving tag detection are discussed. They are named multi-tag and multi-reader. The tag detection of single-tag, two-tag and three-tag are also tested by experiment. The feature of multi-reader is analyzed. They provide a good way for design and applications of RFID system.