arXiv subjects

Xiaowei Tong

Publications and source records attributed to Xiaowei Tong.

5 recordsLinked to original sources

arXiv2026

Counting Trees from Satellite Imagery with Noisy Supervision

Counting individual trees is a fundamental task for environmental monitoring, yet remains largely unexplored with satellite imagery. At these resolutions, isolated trees may still be identifiable, but crown boundaries become ambiguous in dense forests, making the notion of an individual tree inherently ill-defined. Moreover, large-scale manual annotations of individual trees are prohibitively expensive. While scalable supervision can be derived from airborne LiDAR, the resulting annotations are noisy and difficult to exploit effectively. We address these challenges by formulating tree counting as a spatial density matching problem supervised through Unbalanced Optimal Transport. This formulation naturally accommodates both precise localization of isolate trees and robust density estimation in dense forests. We further introduce a self-correction mechanism that leverages transport residuals to progressively refine noisy supervision during training. We evaluate our approach on TinyTrees, a new benchmark spanning three continents and three satellite sensors, comprising over 216 million tree annotations (including 639k manually verified instances) across $25\,890$ km$^2$. Our method consistently outperforms detection-based, regression-based, and transport-based distribution-matching baselines, demonstrating the effectiveness of unbalanced transport and reliability-aware supervision for large-scale tree counting from satellite imagery. Code, data and models are available at https://github.com/dgominski/treematch.

cs.CV↗

arXiv2026

Topological Sensing in the Dynamics of Quantum Walks with Defects

Topological quantum sensing leverages unique topological features to suppress noise and improve the precision of parameter estimation, emerging as a promising tool in both fundamental research and practical application. In this Letter, we propose a sensing protocol that exploits the dynamics of topological quantum walks incorporating localized defects. Unlike conventional schemes that rely on topological protection to suppress disorder and defects, our protocol harnesses the evolution time as a resource to enable precise estimation of the defect parameter. By utilizing topologically nontrivial properties of the quantum walks, the sensing precision can approach the Heisenberg limit. We further demonstrate the performance and robustness of the protocol through Bayesian estimation. Our results show that this approach maintains high precision over a broad range of parameters and exhibits strong robustness against disorder, offering a practical pathway for topologically enhanced quantum metrology.

quant-ph↗

arXiv2023

Mega-cities dominate China's urban greening

Trees play a crucial role in urban environments, offering various ecosystem services that contribute to public health and human well-being. China has initiated a range of urban greening policies over the past decades, however, monitoring their impact on urban tree dynamics at a national scale has proven challenging. In this study, we deployed nano-satellites to quantify urban tree coverage in all major Chinese cities larger than 50 km2 in 2010 and 2019. Our findings indicate that approximately 6000 km2 (11%) of urban areas were covered by trees in 2019, and 76% of these cities experienced an increase in tree cover compared to 2010. Notably, the increase in tree cover in mega-cities such as Beijing, and Shanghai was approximately twice as large as in most other cities (7.69% vs 3.94%). The study employs a data-driven approach towards assessing urban tree cover changes in relation to greening policies, showing clear signs of tree cover increases but also suggesting an uneven implementation primarily benefiting a few mega-cities.

cs.CV↗

arXiv2022

Eco-engineering controls vegetation trends in southwest China karst

The karst area in Yunnan-Guangxi-Guizhou region in southwest China is known for widespread rocky desertification but several studies report a greening trend since the year 2000. While the start of the greening trend seems to match with the implementation of ecological conservation projects, no statistical evidence on a relationship between vegetation greening and eco-engineering exists. Moreover, dominant factors influencing the spatial patterns of vegetation trends have rarely been investigated. Here we use six comprehensive factors representing the natural conditions and human activities of the study area, and several statistical models consistently show that eco-engineering explains large parts of the positive vegetation trends in the karst areas, while negative vegetation trends in non-karst areas of Yunnan were related with a decrease in rainfall. We further show that the interaction of eco-engineering with other factors leads to a heterogeneous pattern of different vegetation trends. Knowing and understanding these patterns is crucial when planning ecological restoration, especially in diverse landscapes like China karst and the methods can be reused in other restoration areas.

physics.app-ph↗

arXiv2018

One "Ruler" for All Languages: Multi-Lingual Dialogue Evaluation with Adversarial Multi-Task Learning

Automatic evaluating the performance of Open-domain dialogue system is a challenging problem. Recent work in neural network-based metrics has shown promising opportunities for automatic dialogue evaluation. However, existing methods mainly focus on monolingual evaluation, in which the trained metric is not flexible enough to transfer across different languages. To address this issue, we propose an adversarial multi-task neural metric (ADVMT) for multi-lingual dialogue evaluation, with shared feature extraction across languages. We evaluate the proposed model in two different languages. Experiments show that the adversarial multi-task neural metric achieves a high correlation with human annotation, which yields better performance than monolingual ones and various existing metrics.

cs.CL↗