摘要 :
This paper reviews the NTIRE 2022 challenge on efficient single image super-resolution with focus on the proposed solutions and results. The task of the challenge was to super-resolve an input image with a magnification factor of ...
展开
This paper reviews the NTIRE 2022 challenge on efficient single image super-resolution with focus on the proposed solutions and results. The task of the challenge was to super-resolve an input image with a magnification factor of ×4 based on pairs of low and corresponding high resolution images. The aim was to design a network for single image super-resolution that achieved improvement of efficiency measured according to several metrics including runtime, parameters, FLOPs, activations, and memory consumption while at least maintaining the PSNR of 29.00dB on DIV2K validation set. IMDN is set as the baseline for efficiency measurement. The challenge had 3 tracks including the main track (runtime), sub-track one (model complexity), and sub-track two (overall performance). In the main track, the practical runtime performance of the submissions was evaluated. The rank of the teams were determined directly by the absolute value of the average runtime on the validation set and test set. In sub-track one, the number of parameters and FLOPs were considered. And the individual rankings of the two metrics were summed up to determine a final ranking in this track. In sub-track two, all of the five metrics mentioned in the description of the challenge including runtime, parameter count, FLOPs, activations, and memory consumption were considered. Similar to sub-track one, the rankings of five metrics were summed up to determine a final ranking. The challenge had 303 registered participants, and 43 teams made valid submissions. They gauge the state-of-the-art in efficient single image super-resolution.
收起
摘要 :
Researchers spend considerable time searching for relevant papers on the topic in which they are currently interested. Often, despite having similar interests, researchers in the same laboratory do not find it convenient to share ...
展开
Researchers spend considerable time searching for relevant papers on the topic in which they are currently interested. Often, despite having similar interests, researchers in the same laboratory do not find it convenient to share results of bibliographic searches and thus conduct independent time-consuming searches. Research paper recommender systems can help the researcher avoid such time-consuming searches by allowing each researcher to automatically take advantage of previous searches performed by others in the lab. Existing recommender systems were developed for commercial domains to assist users by focusing toward products of their interests. Unlike those domains, the research paper domain has relatively few users when compared with the significantly larger number of research papers. In this paper, we present a novel system to recommend relevant research papers to a user based on the user's recent querying and browsing habits. The core of the system is a scalable subspace clustering algorithm, SCuBA (Subspace ClUstering Based Analysis) that performs well on the sparse, high-dimensional data collected in this domain. Both synthetic and benchmark datasets are used to evaluate the recommendation system and to demonstrate that it performs better than the traditional collaborative filtering approaches when recommending research papers.
收起
摘要 :
Straight bar refiner plates is a widely used type of plate for the disc refiner and the dam is the main structure of the plates. Its number and distribution will affect the refining quality and efficiency. In this paper, six strai...
展开
Straight bar refiner plates is a widely used type of plate for the disc refiner and the dam is the main structure of the plates. Its number and distribution will affect the refining quality and efficiency. In this paper, six straight bar refiner plates with different dam number were designed and the effect of dam number on the pulp flow in the refining zone during low consistency pulp refining were investigated by the method of Computational Fluid Dynamics using Fluent software. It showed that pressure decrease from the internal to external radius and the pressure obviously increase at the bar crossing point. And the pressure in the refining zone increases with the number of dams, which provide more compression on fibers, thus more fibrillation could be obtained. While it should be noted that more dams mean more resistance for pulp flow in refining zone. It was shown that the effect of dams on pulp flow velocity is partly, and it slows down the pulp radial velocity near dams and increase pulp tangential velocity. Meanwhile, the pulp residence time in refining zone will be increased due to existence of dams, but low velocity of pulp may block the groove which need to be considered when designing the refiner plates and refining system.
收起
摘要 :
To improve magnetic disturbance rejection and robustness of wellbore survey measurements, an adaptive neuro network-based fuzzy inference system (ANFIS) filter for wellbore position calculation is presented. This technique signifi...
展开
To improve magnetic disturbance rejection and robustness of wellbore survey measurements, an adaptive neuro network-based fuzzy inference system (ANFIS) filter for wellbore position calculation is presented. This technique significantly improves magnetic disturbance rejection and reduces sensor error influence for borehole survey measurements. The new approach for the ANFIS filter is based on two redundant sets of IMUs which are located in different positions in the BHA at a known, constant distance. The distance between these two sets of IMUs will physically fade the effect of the magnetic disturbances. Each IMU set outputs position estimation based on the splines method which is then input into an ANFIS filter. The inputs of the splines calculation are azimuth, inclination angles and measurement depth, and the outputs are moving distance in three directions (Northing, Easting and True Vertical Depth). However, the accuracy of the splines method highly depends on the accuracy of the inputs, which are difficult to obtain during the measurement while drilling process even under pure clean environments (without any magnetic disturbances). Furthermore, the distorted azimuth caused by magnetic interference affects the borehole position accuracy. In order to deal with those problems, the designed ANFIS filter has a two-level structure. First a local level position estimation (splines method or well trained local ANFIS based on the sensor accuracy) for two sensor sets is used. If the sensor measurement accuracy is low, this local ANFIS will correct the position estimation. Then the outputs of the local modules were input into ANFIS for second level filtering (global filter) to remove the error which caused by unknown magnetic disturbances. According to the judgement of the ANFIS, the IMU set with the smaller magnetic disturbance is given greater weight to reduce the interference effect on the borehole position estimation. This two-level filter is compared to the traditional splines method under different tests situations. First, we evaluate this method by comparing with GPS positioning, from this test we know that the ANFIS filter shows a good performance when the magnitude of magnetic disturbance is within the training magnitude range. Even when the magnitude of magnetic disturbance is above the training range, the ANFIS filter shows a higher robustness than the traditional splines method. Also, this method was applied to borehole data with two IMU containing accelerometers and one magnetometer measurements. In order to apply our method, we duplicated one more magnetometer measurement data under magnetic interference for assessment. The results proved its magnetic disturbance robustness in borehole position estimation. Finally, we demonstrate the full potential using a laboratory experimental setup.
收起
摘要 :
This work reviews the results of the NTIRE 2021 Challenge on Non-Homogeneous Dehazing. The proposed techniques and their results have been evaluated on a novel dataset that extends the NH-Haze datset. It consists of additional 35 ...
展开
This work reviews the results of the NTIRE 2021 Challenge on Non-Homogeneous Dehazing. The proposed techniques and their results have been evaluated on a novel dataset that extends the NH-Haze datset. It consists of additional 35 pairs of real haze free and nonhomogeneous hazy images recorded outdoor. The nonhomogeneous haze has been introduced in the outdoor scenes by using a a professional setup that imitates the real conditions of haze scenes. 327 participants registered in the challenge and 23 teams competed in the final testing phase. The proposed solutions gauge the state-of-the-art in image dehazing.
收起
摘要 :
This work reviews the results of the NTIRE 2021 Challenge on Non-Homogeneous Dehazing. The proposed techniques and their results have been evaluated on a novel dataset that extends the NH-Haze datset. It consists of additional 35 ...
展开
This work reviews the results of the NTIRE 2021 Challenge on Non-Homogeneous Dehazing. The proposed techniques and their results have been evaluated on a novel dataset that extends the NH-Haze datset. It consists of additional 35 pairs of real haze free and nonhomogeneous hazy images recorded outdoor. The nonhomogeneous haze has been introduced in the outdoor scenes by using a a professional setup that imitates the real conditions of haze scenes. 327 participants registered in the challenge and 23 teams competed in the final testing phase. The proposed solutions gauge the state-of-the-art in image dehazing.
收起
摘要 :
Community detection is an unsupervised learning task that discovers groups such that group members share more similarities or interact more frequently among themselves than with people outside groups. In social media, link informa...
展开
Community detection is an unsupervised learning task that discovers groups such that group members share more similarities or interact more frequently among themselves than with people outside groups. In social media, link information can reveal heterogeneous relationships of various strengths, but often can be noisy. Since different sources of data in social media can provide complementary information, e.g., bookmarking and tagging data indicates user interests, frequency of commenting suggests the strength of ties, etc., we propose to integrate social media data of multiple types for improving the performance of community detection. We present a joint optimization framework to integrate multiple data sources for community detection. Empirical evaluation on both synthetic data and real-world social media data shows significant performance improvement of the proposed approach. This work elaborates the need for and challenges of multi-source integration of heterogeneous data types, and provides a principled way of multi-source community detection.
收起
摘要 :
Community detection is an unsupervised learning task that discovers groups such that group members share more similarities or interact more frequently among themselves than with people outside groups. In social media, link informa...
展开
Community detection is an unsupervised learning task that discovers groups such that group members share more similarities or interact more frequently among themselves than with people outside groups. In social media, link information can reveal heterogeneous relationships of various strengths, but often can be noisy. Since different sources of data in social media can provide complementary information, e.g., bookmarking and tagging data indicates user interests, frequency of commenting suggests the strength of ties, etc., we propose to integrate social media data of multiple types for improving the performance of community detection. We present a joint optimization framework to integrate multiple data sources for community detection. Empirical evaluation on both synthetic data and real-world social media data shows significant performance improvement of the proposed approach. This work elaborates the need for and challenges of multi-source integration of heterogeneous data types, and provides a principled way of multi-source community detection.
收起
摘要 :
Graphs present in many real-world applications, such as financial fraud detection, commercial recommendation, and social network analysis. But given the high cost of graph annotation or labeling, we face a severe graph label-scarc...
展开
Graphs present in many real-world applications, such as financial fraud detection, commercial recommendation, and social network analysis. But given the high cost of graph annotation or labeling, we face a severe graph label-scarcity problem, i.e., a graph might have a few labeled nodes. One example of such a problem is the so-called few-shot node classification. A predominant approach to this problem resorts to episodic meta-learning. In this work, we challenge the status quo by asking a fundamental question whether meta-learning is a must for few-shot node classification tasks. We propose a new and simple framework under the standard few-shot node classification setting as an alternative to meta-learning to learn an effective graph encoder. The framework consists of supervised graph contrastive learning with novel mechanisms for data augmentation, subgraph encoding, and multi-scale contrast on graphs. Extensive experiments on three benchmark datasets (CoraFull, Reddit, Ogbn) show that the new framework significantly outperforms state-of-the-art meta-learning based methods.
收起
摘要 :
The effects of different heavy metal ions (Cu~(2+), Ni~(2+), Cd~(2+) and Co~(2+)) on the growth and extracellular polymer substances (EPS) composition for both the metal ions- adapted and unadapted Archaea strain Acidianus manzaen...
展开
The effects of different heavy metal ions (Cu~(2+), Ni~(2+), Cd~(2+) and Co~(2+)) on the growth and extracellular polymer substances (EPS) composition for both the metal ions- adapted and unadapted Archaea strain Acidianus manzaensis grown on pyrite were comparatively studied. Results showed that the tolerance to different heavy metal ions of the adapted strain was significantly different from that of the unadapted strain. For the unadapted strain the tolerance capacities to the heavy metal ions followed the order: Co~(2+)> Cu~(2+)≈Ni~(2+)≈Co~(2+) , while after adaption to each metal ions, the order of the tolerance capacities changed to: Cu~(2+)≈ Ni~(2+)> Co~(2+) > Cd~(2+). The contents of proteins, polysaccharides, and uronic acids in capsule layer EPS and slime layer EPS of both the adapted strains and the unadapted strain at logarithmic and stationary growth phases were investigated, respectively. Results showed that the contents of polysaccharides and proteins of adapted strains were significantly lower in capsule layer, while significantly higher in slime layer than that of unadapted strain at both logarithmic and stationary growth phases. The uronic acids contents of the adapted strains were significantly higher in slime layer than that of unadapted strain at both growth phases. While, in the capsule layer, the uronic acids contents of the adapted strains were slightly higher in capsule layer than that of unadapted strain at logarithmic growth phases but significantly lower at stationary growth phases. By combining with the growth of A. manzaensis, it suggests that though the adapted strains had different resistant capacities to heavy metals ions, the strains might have similar resistance and adaption mechanism to heavy metals.
收起
原文获取