arXiv subjects

Xuanting Hao

Publications and source records attributed to Xuanting Hao.

3 recordsLinked to original sources

arXiv2026

Fate of Secondary Droplets Produced by High-speed Raindrops Interacting with a Liquid Pool

Secondary droplets produced by interactions between falling fluid drops and a liquid pool play a significant role in engineering applications and geophysical processes in nature. This study uses direct numerical simulations to investigate the dynamics of secondary droplets generated by raindrop-liquid pool interactions. The raindrop parameters feature a realistic speed of 7 m/s, effective diameters of 1-4 mm, and surface tension values ranging from $25\%$ to twice the typical air-water interface value. The numerical configurations include both a single raindrop and two raindrops separated by distances between two and four times the raindrop diameter. The secondary droplet size distribution, $N_d$, is found to scale with the droplet radius, $r_s$, as $N_d(r_s)\propto r_s^{-5/2}$, with additional dependencies on surface tension and raindrop diameter. When normalized according to this new scaling law, the droplet size distribution obtained from simulations with different parameter values collapses onto a single curve. Analysis of the impact morphology reveals distinct stages of raindrop interactions and identifies the formation and breakup of a central liquid film. Spatial and temporal analyses of the secondary droplets show that raindrop interaction can influence both the percentage of droplets captured by the cavity and the duration over which they re-merge with the pool. These behaviors arise from the combined effects of differences in the birth times of secondary droplets of various sizes and aerodynamic forcing associated with the cavity airflow.

physics.flu-dyn↗

arXiv2024

Masked Image Modeling Boosting Semi-Supervised Semantic Segmentation

In view of the fact that semi- and self-supervised learning share a fundamental principle, effectively modeling knowledge from unlabeled data, various semi-supervised semantic segmentation methods have integrated representative self-supervised learning paradigms for further regularization. However, the potential of the state-of-the-art generative self-supervised paradigm, masked image modeling, has been scarcely studied. This paradigm learns the knowledge through establishing connections between the masked and visible parts of masked image, during the pixel reconstruction process. By inheriting and extending this insight, we successfully leverage masked image modeling to boost semi-supervised semantic segmentation. Specifically, we introduce a novel class-wise masked image modeling that independently reconstructs different image regions according to their respective classes. In this way, the mask-induced connections are established within each class, mitigating the semantic confusion that arises from plainly reconstructing images in basic masked image modeling. To strengthen these intra-class connections, we further develop a feature aggregation strategy that minimizes the distances between features corresponding to the masked and visible parts within the same class. Additionally, in semantic space, we explore the application of masked image modeling to enhance regularization. Extensive experiments conducted on well-known benchmarks demonstrate that our approach achieves state-of-the-art performance. The code will be available at https://github.com/haoxt/S4MIM.

cs.CV↗

arXiv2021

Parameter Decoupling Strategy for Semi-supervised 3D Left Atrium Segmentation

Consistency training has proven to be an advanced semi-supervised framework and achieved promising results in medical image segmentation tasks through enforcing an invariance of the predictions over different views of the inputs. However, with the iterative updating of model parameters, the models would tend to reach a coupled state and eventually lose the ability to exploit unlabeled data. To address the issue, we present a novel semi-supervised segmentation model based on parameter decoupling strategy to encourage consistent predictions from diverse views. Specifically, we first adopt a two-branch network to simultaneously produce predictions for each image. During the training process, we decouple the two prediction branch parameters by quadratic cosine distance to construct different views in latent space. Based on this, the feature extractor is constrained to encourage the consistency of probability maps generated by classifiers under diversified features. In the overall training process, the parameters of feature extractor and classifiers are updated alternately by consistency regularization operation and decoupling operation to gradually improve the generalization performance of the model. Our method has achieved a competitive result over the state-of-the-art semi-supervised methods on the Atrial Segmentation Challenge dataset, demonstrating the effectiveness of our framework. Code is available at https://github.com/BX0903/PDC.

cs.CV↗