CDS&E: D3SC: Applying Video Segmentation to Coarse-grain Mapping Operators in Molecular Simulations

CDS

• 批准号: 1764415
• 负责人: Andrew White
• 金额: $ 48.86万
• 依托单位: University of Rochester
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2018
• 资助国家: 美国
• 起止时间: 2018-08-01 至 2022-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1764415&HistoricalAwards=false
• 关键词: CDS& D3SC; Applying; Video; Segmentation

Andrew White and Chenliang Xu of the University of Rochester is supported by an award from the Chemical Theory, Models and Computational Methods program in the Division of Chemistry to apply advances in computer vision to improve models of multiscale systems in chemistry. Multiscale systems describe chemical and physical processes that occur on many different time and spatial scales, for example, both very fast and very slow motions may contribute to the overall process. In both the computer processing of videos and the modeling of multiscale chemical systems, reducing complexity via removing extraneous details is essential. Without removing some model details, simulating multiscale processes like DNA transcription or the peptide aggregation which leads to plaque formation in Alzheimer's disease is impossible. Current approaches to reduce the number of atoms in a model rely on intuition and tradition due to the near infinite ways in which atoms can be removed or combined. White, Xu and their research groups are developing a novel approach built upon advances in video segmentation. Video segmentation is the process of identifying foreground, background, and objects in a video. Surprisingly, the same mathematical structure can be applied to chemical systems and that is the goal of this research. White, Xu and their collaborators will introduce the research to a broader audience via an augmented-reality laboratory for students. Students will be able to decide how to simplify molecular models and see the results by combining the visual experience of augmented-reality with the interactivity of molecular simulations. Coarse-graining (CG) is the dimension reduction technique used to simulate multiscale systems more efficiently. There is not a rigorous theory for generating mappings from all-atom (fine-grain) system to the CG system. This missing component is essential because past CG work shows that many mappings lead to homogeneous, weakly interacting, gas-like CG models but the number of possible mappings is combinatorial with respect to the number of atoms. Andrew White and his collaborators are working to solve this mapping problem by (i) developing a theory to represent mapping algorithms based on video segmentation algorithms; (ii) creating a database of mappings and their performance on benchmark simulations to foster community involvement; (iii) studying and testing these methods on multi-protein surface interactions, where current mapping approaches struggle. Achieving success here, along with recent advances in calculating CG potentials, will better advance the community's ability to model complex multiscale phenomena.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

罗切斯特大学的Andrew白色和Chenliang Xu获得了化学系化学理论，模型和计算方法项目的奖项，以应用计算机视觉的进步来改进化学中多尺度系统的模型。 多尺度系统描述了在许多不同的时间和空间尺度上发生的化学和物理过程，例如，非常快和非常慢的运动都可能有助于整个过程。 在视频的计算机处理和多尺度化学系统的建模中，通过去除无关的细节来降低复杂性是必不可少的。如果不去除一些模型细节，模拟多尺度过程，如DNA转录或导致阿尔茨海默病斑块形成的肽聚集是不可能的。目前减少模型中原子数量的方法依赖于直觉和传统，因为原子可以被移除或组合的方式几乎是无限的。白色，徐和他们的研究小组正在开发一种新的方法，建立在视频分割的进步。视频分割是识别视频中的前景、背景和对象的过程。令人惊讶的是，相同的数学结构可以应用于化学系统，这就是本研究的目标。白色，徐和他们的合作者将通过一个面向学生的增强现实实验室向更广泛的受众介绍这项研究。学生将能够决定如何简化分子模型，并通过将增强现实的视觉体验与分子模拟的交互性相结合来查看结果。粗粒化（CG）是一种用于更有效地模拟多尺度系统的降维技术。从全原子（细颗粒）系统到CG系统的映射还没有一个严格的理论。这个缺失的部分是必不可少的，因为过去的CG工作表明，许多映射导致均匀的，弱相互作用的，气体样的CG模型，但可能的映射的数量是组合相对于原子的数量。Andrew白色和他的合作者正在努力解决这个映射问题，方法是：（i）开发一种理论来表示基于视频分割算法的映射算法;（ii）创建一个映射数据库及其在基准模拟上的性能，以促进社区参与;（iii）在多蛋白质表面相互作用上研究和测试这些方法，目前的映射方法正在努力。在这里取得成功，沿着计算CG潜力的最新进展，将更好地提高社区模拟复杂多尺度现象的能力。该奖项反映了NSF的法定使命，并被认为值得通过使用基金会的智力价值和更广泛的影响审查标准进行评估来支持。

项目成果

Audio-Visual Interpretable and Controllable Video Captioning

• 发表时间: 2019
• 作者: Yapeng Tian;Chenxiao Guan;Justin Goodman;Marc Moore;Chenliang Xu

Discover and Mitigate Unknown Biases with Debiasing Alternate Networks

• DOI: 10.48550/arxiv.2207.10077
• 发表时间: 2022-07
• 作者: Zhiheng Li;A. Hoogs;Chenliang Xu

GAN-EM: GAN based EM learning framework

• DOI: 10.24963/ijcai.2019/612
• 发表时间: 2018-12
• 作者: Wentian Zhao;Shaojie Wang;Zhihuai Xie;Jing Shi;Chenliang Xu

Discover the Unknown Biased Attribute of an Image Classifier

• DOI: 10.1109/iccv48922.2021.01470
• 发表时间: 2021-04
• 期刊: 2021 IEEE/CVF International Conference on Computer Vision (ICCV)
• 作者: Zhiheng Li;Chenliang Xu

Sound to Visual: Hierarchical Cross-Modal Talking Face Video Generation

• 发表时间: 2019
• 作者: Lele Chen;Haitian Zheng;Ross K Maddox;Zhiyao Duan;Chenliang Xu