arXiv subjects

Yinan Yang

Publications and source records attributed to Yinan Yang.

3 recordsLinked to original sources

arXiv2023

Development of a Chemistry Dynamic Load Balancing Solver with Sparse Analytical Jacobian Approach for Rapid and Accurate Reactive Flow Simulations

In addressing the demands of industrial high-fidelity computation, the present study introduces a rapid and accurate customized solver developed on the OpenFOAM platform. To enhance computational efficiency, a novel integrated acceleration strategy is introduced. Initially, a sparse analytical Jacobian approach utilizing the SpeedCHEM chemistry library was implemented to increase the efficiency of the ODE solver. Subsequently, the Dynamic Load Balancing (DLB) code was employed to uniformly distribute the computational workload for chemistry among multiple processes. Further optimization was achieved through the introduction of the Open Multi-Processing (OpenMP) method to enhance parallel computing efficiency. Lastly, the Local Time Stepping (LTS) scheme was integrated to maximize the individual time step for each computational cell, resulting in a noteworthy minimum speed-up of over 31 times. The effectiveness and robustness of this customized solver were systematically validated against three distinct partially turbulent premixed flames, Sandia Flames D, E, and F. Additionally, a comparative analysis was conducted, encompassing different turbulence models, turbulent Prandtl numbers, and model constants, resulting in the recommendation of optimal numerical parameters for various conditions. The present study offers one viable solution for rapid and accurate calculations in the OpenFOAM platform, while also providing insights into the selection of turbulence models and parameters for industrial numerical simulation.

physics.flu-dyn↗

arXiv2022

WebUAV-3M: A Benchmark for Unveiling the Power of Million-Scale Deep UAV Tracking

Unmanned aerial vehicle (UAV) tracking is of great significance for a wide range of applications, such as delivery and agriculture. Previous benchmarks in this area mainly focused on small-scale tracking problems while ignoring the amounts of data, types of data modalities, diversities of target categories and scenarios, and evaluation protocols involved, greatly hiding the massive power of deep UAV tracking. In this work, we propose WebUAV-3M, the largest public UAV tracking benchmark to date, to facilitate both the development and evaluation of deep UAV trackers. WebUAV-3M contains over 3.3 million frames across 4,500 videos and offers 223 highly diverse target categories. Each video is densely annotated with bounding boxes by an efficient and scalable semiautomatic target annotation (SATA) pipeline. Importantly, to take advantage of the complementary superiority of language and audio, we enrich WebUAV-3M by innovatively providing both natural language specifications and audio descriptions. We believe that such additions will greatly boost future research in terms of exploring language features and audio cues for multimodal UAV tracking. In addition, a fine-grained UAV tracking-under-scenario constraint (UTUSC) evaluation protocol and seven challenging scenario subtest sets are constructed to enable the community to develop, adapt and evaluate various types of advanced trackers. We provide extensive evaluations and detailed analyses of 43 representative trackers and envision future research directions in the field of deep UAV tracking and beyond. The dataset, toolkits and baseline results are available at \url{https://github.com/983632847/WebUAV-3M}.

cs.CV↗

arXiv2022

One-shot Network Pruning at Initialization with Discriminative Image Patches

One-shot Network Pruning at Initialization (OPaI) is an effective method to decrease network pruning costs. Recently, there is a growing belief that data is unnecessary in OPaI. However, we obtain an opposite conclusion by ablation experiments in two representative OPaI methods, SNIP and GraSP. Specifically, we find that informative data is crucial to enhancing pruning performance. In this paper, we propose two novel methods, Discriminative One-shot Network Pruning (DOP) and Super Stitching, to prune the network by high-level visual discriminative image patches. Our contributions are as follows. (1) Extensive experiments reveal that OPaI is data-dependent. (2) Super Stitching performs significantly better than the original OPaI method on benchmark ImageNet, especially in a highly compressed model.

cs.CV↗