Search

Your search keyword '"Chen, Songming"' showing total 124 results
Start Over Author "Chen, Songming"
124 results on '"Chen, Songming"'

1. PrevPredMap: Exploring Temporal Modeling with Previous Predictions for Online Vectorized HD Map Construction

Author

Peng, Nan, Zhou, Xun, Wang, Mingming, Yang, Xiaojun, Chen, Songming, and Chen, Guisong

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Temporal information is crucial for detecting occluded instances. Existing temporal representations have progressed from BEV or PV features to more compact query features. Compared to these aforementioned features, predictions offer the highest level of abstraction, providing explicit information. In the context of online vectorized HD map construction, this unique characteristic of predictions is potentially advantageous for long-term temporal modeling and the integration of map priors. This paper introduces PrevPredMap, a pioneering temporal modeling framework that leverages previous predictions for constructing online vectorized HD maps. We have meticulously crafted two essential modules for PrevPredMap: the previous-predictions-based query generator and the dynamic-position-query decoder. Specifically, the previous-predictions-based query generator is designed to separately encode different types of information from previous predictions, which are then effectively utilized by the dynamic-position-query decoder to generate current predictions. Furthermore, we have developed a dual-mode strategy to ensure PrevPredMap's robust performance across both single-frame and temporal modes. Extensive experiments demonstrate that PrevPredMap achieves state-of-the-art performance on the nuScenes and Argoverse2 datasets. Code will be available at https://github.com/pnnnnnnn/PrevPredMap.

Published
2024

12. Reliability assessment of degradation processes with random shocks considering recoverable shock damages

Author

Huang, Tingting, Chen, Songming, Zhao, Yuepu, and Dai, Wei

Abstract

For the products that are exposed to random shocks in the field use, the continuous degradation processes of the products are often affected by random shocks. There are two types of shock effects typically: (1) a sudden change of the degradation signal; (2) an increment of the degradation rate. For some kinds of the products, such as the material with recovery properties and smart systems with self-recovery properties, the sudden change of degradation signals caused by shocks can be fully or partially recovered after the occurrence of the shocks. Thus, the recoverable shock damage should be considered in the degradation modeling and reliability prediction of such products. In this paper, a degradation model considering recoverable shock damage is established based on a Wiener process. Three transition zones are divided according to different shock effects. The arrival of the random shocks is expressed by a homogeneous Poisson process, and the shock magnitudes are non-negative independent random variables following a normal distribution. The shock effect is described as the cumulative shock damage function consisting of both the effect on the degradation signal which is expressed by an exponential function with/without a residual effect for a partial or full recovery, and on the degradation rate which is proportional to the shock magnitude when the shock magnitude exceeds the critical threshold. Parameter estimation and reliability prediction are discussed. Numerical examples are presented as demonstrations of the applications of the proposed model.

Published
2023
Full Text
View/download PDF

13. Perception et estimation d’état basées sur plusieurs capteurs pour les véhicules autonomes

Author

Chen, Songming, Laboratoire des Sciences du Numérique de Nantes (LS2N), Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-IMT Atlantique (IMT Atlantique), Institut Mines-Télécom [Paris] (IMT)-Institut Mines-Télécom [Paris] (IMT)-École Centrale de Nantes (Nantes Univ - ECN), Nantes Université (Nantes Univ)-Nantes Université (Nantes Univ)-Nantes université - UFR des Sciences et des Techniques (Nantes univ - UFR ST), Nantes Université - pôle Sciences et technologie, Nantes Université (Nantes Univ)-Nantes Université (Nantes Univ)-Nantes Université - pôle Sciences et technologie, Nantes Université (Nantes Univ), École centrale de Nantes, and Vincent Frémont

Subjects
Sensor fusion, Environment perception, Autonomous vehicles, Uncertainty analysis, Analyse d’incertitude, Véhicules autonomes, Perception de l’environnement, Estimation d’état, State estimation, Fusion de capteurs, [SPI.AUTO]Engineering Sciences [physics]/Automatic
Abstract

Perceiving or understanding the surrounding environments is indispensable for building driving assistant systems or autonomousvehicles. This thesis studies the sensor fusion approach for the simultaneous localization and mapping (SLAM) problem with complementary visual and range sensors. In order to make conservative decisions and increase autonomous vehicle maneuveringsecurity, the uncertainty analysis of the pose estimation is also implemented. The traditional SLAM system has the assumption of static scenes, which is vulnerable in the context of dynamic outdoor environments. Thus, we introduce a data-driven approach to exploit the semantic information that interprets the measurement sequence across the frames, which efficiently distinguishes the moving objects from the static ones. We test the proposed algorithms on real-life data of urban traffic and parking areas, which presents promising results.; Percevoir ou comprendre les environnements environnants est indispensable pour construire des systèmes d’aide à la conduite ou des véhicules autonomes. Dans cette thèse, nous étudions l’approche de fusion de capteurs pour le problème de localisation et de cartographie simultanées (SLAM) avec des capteurs visuels et de distance complémentaires. Afin de prendre des décisionsconservatrices et d’augmenter la sécurité de manoeuvre des véhicules autonomes, l’analyse d’incertitude de l’estimation de la pose est également mise en oeuvre. Le système SLAM traditionnel suppose des scènes statiques, ce qui est vulnérable dans le contexte d’environnements extérieurs dynamiques. Ainsi, nous introduisons une approche basée sur les données pour exploiter les informations sémantiques qui interprètent la séquence de mesure à travers les cadres, ce qui distingue efficacement les objets en mouvement des objets statiques. Nous testons les algorithmes proposé sur des données réelles de trafic urbain et d’aires de stationnement, qui présente des résultats prometteurs.

Published
2022

14. A Semantic-Guided LiDAR-Vision Fusion Approach for Moving Objects Segmentation and State Estimation

Author

Chen, Songming, Sun, Haixin, Frémont, Vincent, Laboratoire des Sciences du Numérique de Nantes (LS2N), Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-IMT Atlantique (IMT Atlantique), Institut Mines-Télécom [Paris] (IMT)-Institut Mines-Télécom [Paris] (IMT)-École Centrale de Nantes (Nantes Univ - ECN), Nantes Université (Nantes Univ)-Nantes Université (Nantes Univ)-Nantes université - UFR des Sciences et des Techniques (Nantes univ - UFR ST), Nantes Université - pôle Sciences et technologie, Nantes Université (Nantes Univ)-Nantes Université (Nantes Univ)-Nantes Université - pôle Sciences et technologie, Nantes Université (Nantes Univ), Autonomie des Robots et Maîtrise des interactions avec l’ENvironnement (LS2N - équipe ARMEN), Nantes Université (Nantes Univ)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-IMT Atlantique (IMT Atlantique), and ANR-16-IDEX-0007,NExT (I-SITE),NExT (I-SITE)(2016)

Subjects
[INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], [SPI.AUTO]Engineering Sciences [physics]/Automatic
Abstract

International audience; Moving Objects Segmentation (MOS) is critical and indispensable for secure intelligent vehicle operation in the dynamic environment. For the state estimation task which is based on the assumption of static surroundings, to identify and filter out the moving objects plays an important role in robust ego-motion estimation. In this paper, a LiDAR-Vision fusion approach is developed to segment moving objects in the scene, which utilizes the LiDAR-based semantic segmentation as a prior and vision-based geometric information for validation. The effectiveness of our approach to segment moving objects is highlighted by the comparison with the traditional robust kernel-based outlier rejection methods. Our approach is benchmarked with three city category sequences in the KITTI dataset, which outperforms the kernel-based methods and achieves the leading results of 77.9% average fitness and 7.65 cm RMSE respectively.

Published
2022

26. Amyloid-like features of polyglutamine aggregates and their assembly kinetics

Author

Chen, Songming, Berthelier, Valerie, Hamilton, J. Bradley, O'Nuallain, Brian, and Wetzel, Ronald

Subjects
Biochemistry -- Research, Amyloid beta-protein -- Physiological aspects, Glutamine -- Physiological aspects, Huntington's chorea -- Causes of, Neurons -- Physiological aspects, Biological sciences, Chemistry
Abstract

Research has been conducted on the polyglutamine sequences. The features and assembly pathways of several polyglutamine aggregates resembling those of amyloid fibrils have been investigated and the results suggest that their ordered structure may be responsible for the observed in Huntington's and other repeat diseases cell-specific neuronal degeneration.

Published
2002

35. A Glycoform of Immunoglobulin G (IgG) as an Early Biomarker of Exposure to Nonhuman Substances

Author

INSTITUTE FOR HEPATITIS AND VIRUS RESEARCH (IHVR) DOYLESTOWN PA, Block, Timothy M, Chen, Songming, Mehta, Anand S, Henderson, Terry J, INSTITUTE FOR HEPATITIS AND VIRUS RESEARCH (IHVR) DOYLESTOWN PA, Block, Timothy M, Chen, Songming, Mehta, Anand S, and Henderson, Terry J

Abstract

We have identified an immunoglobulin G (IgG) subset that transiently exists in rabbits and mice following the introduction of various nonself antigens. This previously unrecognized IgG was named primebody because it appears to be involved in priming the humoral immune response rather than in neutralizing immunogens. Primebody was initially identified by its ability to bind the Aleuria aurantia lectin (AAL) while still in its native conformation, in contrast to serum IgGs that bind AAL poorly, unless denatured. For both animals inoculated with three different immunogens, serum levels of primebody that were investigated with an AAL-antibody microarray assay reached peak levels 5 to 16 days post-inoculation. However, the immunogen-specific IgG levels peaked much later, between 21 to 60 days post-inoculation. Mass spectrometry analysis revealed 22 different primebody glycoisoforms in rabbits, all of which were under-galactosylated IgG (G0 and G1) with core fucose on its constant region N-glycan. We hypothesized that the immunoglobulin constant region heavy chain 2 (CH2) domain of primebody adopts an open conformation that allows AAL complete access to its glycans, in contrast with the usual closed conformation adopted by immunoglobulin CH2 domains, which are known to deny all glycan access., The original document contains color images.

Published
2012

36. Aleuria Aurantia Lectin (AAL)-Reactive Immunoglobulin G Rapidly Appears in Sera of Animals following Antigen Exposure.

Author

Chen, Songming, Lu, Chen, Gu, Hongbo, Mehta, Anand, Li, Jianwei, Romano, Patrick B, Horn, David, Hooper, D Craig, Bazemore-Walker, Carthene R, Block, Timothy, Chen, Songming, Lu, Chen, Gu, Hongbo, Mehta, Anand, Li, Jianwei, Romano, Patrick B, Horn, David, Hooper, D Craig, Bazemore-Walker, Carthene R, and Block, Timothy

Abstract

We have discovered an Aleuria Aurantia Lectin (AAL)-reactive immunoglobulin G (IgG) that naturally occurs in the circulation of rabbits and mice, following immune responses induced by various foreign antigens. AAL can specifically bind to fucose moieties on glycoproteins. However, most serum IgGs are poorly bound by AAL unless they are denatured or treated with glycosidase. In this study, using an immunogen-independent AAL-antibody microarray assay that we developed, we detected AAL-reactive IgG in the sera of all animals that had been immunized 1-2 weeks previously with various immunogens with and without adjuvants and developed immunogen-specific responses. All of these animals subsequently developed immunogen-specific immune responses. The kinetics of the production of AAL-reactive IgG in mice and rabbits were distinct from those of the immunogen-specific IgGs elicited in the same animals: they rose and fell within one to two weeks, and peaked between four to seven days after exposure, while immunogen-specific IgGs continued to rise during the same period. Mass spectrometric profiling of the Fc glycoforms of purified AAL-reactive IgGs indicates that these are mainly comprised of IgGs with core-fucosylated and either mono-or non-galactosylated Fc N-glycan structures. Our results suggest that AAL-reactive IgG could be a previously unrecognized IgG subset that is selectively produced at the onset of a humoral response.

Published
2012

50. Solubilization and disaggregation of polyglutamine peptides.

Author

Chen, Songming and Wetzel, Ronald

Abstract

A method is described for dissolving and disaggregating chemically synthesized polyglutamine peptides. Polyglutamine peptides longer than about Q20 have been reported to be insoluble in water, but dissolution in - and evaporation from - a mixture of trifluoroacetic acid and hexafluoroisopropanol converts polyglutamine peptides up to at least Q44 to a form readily soluble in aqueous buffers. This procedure also has a dramatic effect on peptides which appear to be completely soluble in water, by removing traces of aggregate that seed aggregation. The protocol makes possible solution studies-including in vitro aggregation experiments-on polyglutamine peptides with repeat lengths associated with increased risk of Huntington's Disease and other expanded CAG repeat diseases. It may also be useful in conducting reproducible, quantitative aggregation studies on other polypeptides. [ABSTRACT FROM AUTHOR]

Published
2001
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results