A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is
Vehicle re-identification aims to obtain the same vehicles from vehicle images. This is challenging but essential for analyzing and predicting traffic flow in the city. Although deep learning methods have achieved enormous progress for this task, their large data requirement is a critical shortcoming. Therefore, we propose a synthetic-to-real domain adaptation network (StRDAN) framework, which can be trained with inexpensive large-scale synthetic and real data to improve performance. The StRDANarXiv:2004.12032v2 fatcat:ltiuxfrtlnbs7ekmzgpjbigtgu
more »... training method combines domain adaptation and semi-supervised learning methods and their associated losses. StRDAN offers significant improvement over the baseline model, which can only be trained using real data, for VeRi and CityFlow-ReID datasets, achieving 3.1% and 12.9% improved mean average precision, respectively.
Traffic analysis using computer vision techniques is attracting more attention for the development of intelligent transportation systems. Consequently, counting traffic volume based on the CCTV system is one of the main applications. However, this issue is still a challenging task, especially in the case of complex areas that involve many vehicle movements. This study performs an investigation of how to improve video-based vehicle counting for traffic analysis. Specifically, we propose adoi:10.3390/en13082036 fatcat:mebzbxb7jrep7kkazt5njlvjey
more »... ensive framework with multiple classes and movements for vehicle counting. In particular, we first adopt state-of-the-art deep learning methods for vehicle detection and tracking. Then, an appropriate trajectory approach for monitoring the movements of vehicles using distinguished regions tracking is presented in order to improve the performance of the counting. Regarding the experiment, we collect and pre-process the CCTV data at a complex intersection to evaluate our proposed framework. In particular, the implementation indicates the promising results of our proposed method, which achieve accuracy around 80% to 98% for different movements for a very complex scenario with only a single view of the camera.
We report pseudopotential density-functional calculations of the "3 Â 1" phase of the Ba/Si (111) surface. To resolve the coverage issue of the Ba/Si(111) surface, we investigate various structural models with two Ba coverages: 1/3 ML and 1/6 ML. Based on the comparison of the simulated STM images for two coverages and the experimental images, it is found that the "3 Â 1" phase has 1/6 ML Ba and a 3 Â 2 structure with the honeycomb chain-channel Si reconstruction. The substrate reconstructiondoi:10.1143/jpsj.71.2761 fatcat:6akh35dyknchrd4s7s4h47mroy
more »... quite similar to the Si(111)-3 Â 1 surface induced by 1/3 ML alkali-metals. The experimental semiconducting band gap is well reproduced by the LDA band structure.
The loss to make the network be trained indiscriminative to two domain is defined as follows: L domain = 1 N N i=1 y i log(ŷ i ) + (1 − y i ) log(1 − ŷi ). ( 3 ) The domain discrimination loss is defined ...doi:10.1109/cvprw50498.2020.00312 dblp:conf/cvpr/LeePYL20 fatcat:4cl6rffbv5dn7bpmkjghgg36vy
Journal of KIISE
도심지 교통흐름 및 미세먼지 예측을 위한 딥러닝 LSTM 프레임워크
도심지 교통흐름 및 미세먼지 예측을 위한 딥러닝 LSTM 프레임워크
Yi) (Khac-Hoai Nam Bui) (Choong-Nyoung Seon) -클래스 입출력 구조를 만들어 지도학습 기반의 LSTM 모형을 설계하였다. ... 정보과학회논문지 제47권 제3호(2020. 3) 도심지 교통흐름 및 미세먼지 예측을 위한 딥러닝 LSTM 프레임워크 (A Deep Learning LSTM Framework for Urban Traffic Flow and Fine Dust Prediction) 이 홍 석 † 부이 칵 남 † † 선 충 녕 † † (Hongsuk ...doi:10.5626/jok.2020.47.3.292 fatcat:ofapg2fkn5hndhe6shtdeop7mq
From a Monte Carlo study of the ferromagnetic Kondo lattice model for doped manganites, including the antiferromagnetic superexchange interaction (J_AF), we found that the ferromagnetic ordering was suppressed as J_AF increased. The ferromagnetic transition temperature T_c, as obtained from a mean field fit to the calculated susceptibilities, was found to decrease monotonically with increasing J_AF. Further, the suppression in T_c scales with the bandwidth narrowing induced by thedoi:10.1103/physrevb.61.428 fatcat:okfohn74lndqvgck76hdobfrca
more »... tic frustration originating from J_AF. From these results, we propose that the change in the superexchange interaction strength between the t_2g electrons of the Mn ions is one of the mechanisms responsible for the suppression in T_c observed in manganites of the type (La_0.7-yPr_y)Ca_0.3MnO_3.
The magnetic susceptibility and spin-spin correlation of the double-exchange model for doped manganites are investigated through the Monte Carlo calculations on the three-dimensional lattice model. Deviations of the susceptibility from the Curie-Weiss behavior above the ferromagnetic ordering temperature T_c seem to indicate a formation of local ferromagnetic clusters in the vicinity of T_c, which is consistent with recent electron paramagnetic resonance experiments for La_2/3Ca_1/3MnO_3. Adoi:10.1103/physrevb.61.9501 fatcat:lvi6qndjhjeqjpljp5pdglj5ku
more »... her analysis of the spin-spin correlations show the ferromagnetic cluster size to be three-to-four lattice spacings, suggesting that the charge carriers may form magnetic polarons.
Current Research on Software, Security, Cloud and Super Computing
In order to secure the capability comparable to the advanced countries in supercomputing area, it is necessary to give first priority to manpower training. We surveyed several supercomputing education and training programs such as NSF XSEDE project, EU PRACE project, and etc. We pick out two main activities, development supercomputing contents and manpower certification system, to upgrade our supercomputing manpower in a short period time.doi:10.14257/astl.2013.35.24 fatcat:s2jc23cvpvdchlijg7cgj2srsa
Virginia Medical Quarterly : VMQ
Julius Keith, Midlothian Lesko, Edmund Michael, Rocky Mount Levin, Laurence L, Virginia Beach Li, Si-Ju, Alexandria Lichtmann, Albert Laszlo, McLean Lim, Henry Suh, Norfolk Lin, Allen Y, Danville Lin, Yi-Shien ... Lynchburg Shim, Chi Yun, Richmond Sibay, Hassan, Mc Lean Silbersiepe, Barbara Kromer, Catlett Silbersiepe, Heinz-Otto, Catlett Silbersiepe, Kathryn Ann, Richmond Singh, Harinder P, Richmond Sirh, Joseph Hongsuk ...
., 2002h:62131 (Jing, Bing-Yi) see Kumazawa, Yoshiki Jorn, Hongsuk (with Klotz, Jerome) Exact distribution of the K sample mood and Brown median test. ... (English summary) 2002b:62057 Kuan, Chung-Ming see Chen, Yi-Ting, 2002):62063 Kulasekera, K. B. (with Wang, Jian®) A test of equality of regression curves using Gateaux scores. ...
Losev) 2004¢:35126 35360 (35B40, 58E12) — (with Fang, Yi!) When is a minimal surface a minimal graph? (English summary) Pacific J. Math. 207 (2002), no. 2, 359-376. ... (Zhi Ren Jin) 2004h:53011 53A10 (35B05, 35560) Hwang, Jinsoo (with Jun, Sung-Hae; Jorn, Hongsuk) Bayesian learning for self organizing maps. (Korean. English and Korean summaries) Korean J. Appl. ...
(English summary) James, Lancelot F. see Ishwaran, Hemant, 2004d:62018 Jorn, Hongsuk see Jun, Sung-Hae et al. ... .; Shao, Qi-Man) Propriety of the posterior distribution and existence of the MLE for regression models with covariates missing at random. 2004m:62020 Cheng, Yi? ...
Yi Zhang José Matas Yi Zheng José Miguel Burdío Yi Zhou José ... Mullat Yi-tang Chang Joseph H.K. ...doi:10.3390/en15031171 fatcat:ed5twpuvl5cilkzfjwwzgrk5cm
Physical Review D
Agashe, James Gainer, Ian Lewis, Konstantin Matchev, Frank Paige, Gilad Perez, and Hong Zhang for helpful comments and discussions, Roberto Franceschini for extensive discussions on the fitting functions, Hongsuk ...doi:10.1103/physrevd.89.096007 fatcat:xjskhqiqlfetdnepkotytielze
« Previous Showing results 1 — 15 out of 18 results