Institution profile

Shenzhen University

Academic institutionasia · cn
Official website
Research library1,011linked papers
Opportunities0open roles
Selected work

Representative Papers

Prompting Depth Anything for 4K Resolution Accurate Metric Depth Estimation

Dec 18, 2024arXiv.org

To resolve the accuracy–cost trade-off between low-cost LiDAR and high-resolution metric depth estimation, this paper introduces Prompt Depth Anything—a novel paradigm that leverages sparse, low-accuracy LiDAR point clouds as multi-scale geometric prompts to guide the Depth Anything foundation model toward 4K-resolution metric depth prediction. Methodologically, we design a lightweight prompt fusion architecture enabling cross-scale feature alignment and develop a scalable data pipeline integrating LiDAR physics simulation with pseudo-ground-truth generation from real-world scenes. Evaluated on ARKitScenes and ScanNet++, our approach achieves state-of-the-art performance, reducing 4K depth error by 21.3% relatively. Moreover, the high-fidelity depth maps substantially enhance downstream applications, including photorealistic 3D reconstruction and general-purpose robotic grasping.

8 citationsRead paper

Privacy Protection in Prosumer Energy Management Based on Federated Learning

Mar 09, 2025IEEE Access

To address privacy leakage, degraded model accuracy, and excessive communication overhead caused by Non-IID data in prosumer energy management systems, this paper proposes FedClusAvg, a novel federated learning algorithm. FedClusAvg introduces clustered hierarchical sampling and parameter-deviation-weighted aggregation, operating within a three-tier architecture (central server–sub-server–client) that supports multi-round local training. Crucially, it ensures raw electricity consumption data remains on-device while significantly improving model accuracy under Non-IID conditions and reducing total communication rounds by over 40%. Its core contribution lies in jointly optimizing privacy preservation, modeling fidelity, and system efficiency—delivering a scalable, robust, and privacy-enhancing solution for distributed energy coordination and optimization.

3 citationsRead paper

Dynamic Spectrum Sharing Based on the Rentable NFT Standard ERC4907

May 10, 2024International Conference on Communications, Circuits and Systems

Centralized dynamic spectrum sharing (DSS) faces critical challenges including data security vulnerabilities, high administrative overhead, and poor scalability. To address these issues, this paper proposes the first blockchain-based DSS system built upon the ERC-4907 rentable NFT standard. It introduces Non-Fungible Spectrum Tokens (NFSTs) to uniquely identify spectrum resources while decoupling ownership from usage rights. Leveraging on-chain smart contract–driven auctions and a Web3-enabled frontend platform, the system enables fine-grained, verifiable, and auditable decentralized spectrum leasing. Experimental evaluation demonstrates significant improvements in spectrum allocation security and transaction efficiency, substantial reduction in management costs, and robust support for high-concurrency leasing requests. This work constitutes the first empirical validation of ERC-4907’s feasibility and practicality for wireless spectrum governance.

3 citationsRead paper

Tumor Detection, Segmentation and Classification Challenge on Automated 3D Breast Ultrasound: The TDSC-ABUS Challenge

Jan 26, 2025

Automatic breast ultrasound (ABUS) tumor detection, segmentation, and classification are challenged by morphological heterogeneity, low signal-to-noise ratio, and scarcity of annotated 3D data. Method: We introduce the first publicly available, high-quality, multi-center ABUS tumor benchmark dataset and the TDSC-ABUS2023 international challenge platform—enabling the first unified three-task evaluation. Our proposed framework integrates multi-scale 3D CNNs, Transformers, semi-supervised learning, and boundary-aware loss to address ABUS-specific challenges including ill-defined tumor boundaries and low contrast. Contribution/Results: Our method achieves state-of-the-art performance: 82.3% mAP@0.5 for detection, 79.6% Dice for segmentation, and 91.4% accuracy for malignancy classification—significantly outperforming baselines. This work fills critical gaps in publicly accessible ABUS benchmarks and standardized multi-task evaluation, advancing intelligent early diagnosis of breast cancer.

2 citationsRead paper

InfiniDepth: Arbitrary-Resolution and Fine-Grained Depth Estimation with Neural Implicit Fields

Jan 06, 2026arXiv.org

This work addresses the limitations of conventional depth estimation methods, which are constrained by discrete image grids and thus struggle to recover fine geometric details or support arbitrary-resolution outputs. To overcome this, we propose a continuous depth representation based on neural implicit fields, introducing a local implicit decoder that enables high-fidelity depth querying at any 2D coordinate. To facilitate training and evaluation, we construct a high-resolution 4K synthetic dataset. Experimental results demonstrate that our approach achieves state-of-the-art performance on both synthetic and real-world datasets, significantly enhancing geometric detail recovery and substantially improving the quality of novel-view synthesis under large viewpoint changes.

1 citationsRead paper
Recent publications

Latest Papers