Search

Methods based on 3D Gaussian Splatting (3DGS) for surface reconstruction face challenges when applied to large-scale scenes captured by UAV. Because the number of 3D Gaussians increases dramatically, leading to significant computational requirement and limiting the fineness of surface reconstruction. To address this challenge, we propose C3DGS that compresses 3D Gaussian model and ensures the quality of surface reconstruction of large-scale scenes in the face of heavy computational costs. Our method quantifies the contribution of 3D Gaussians to the surface reconstruction and prunes redundant 3D Gaussians to reduce the computational requirement of the model. In addition, pruning 3D Gaussians inevitably incurs loss, and in order to guarantee as many details as possible in the surface reconstruction of a complex scene, we use a ray tracing volume rendering method that can better evaluate the opacity of 3D Gaussians. Furthermore, we introduce two regularization terms to enhance the geometric consistency of multiple views, thus improving the realism of surface reconstruction. Experiments show that our method outperforms other 3DGS-based surface reconstruction methods when facing large-scale scenes.

Artificial Intelligence (AI) Machine Learning (ML) technologies, particularly Deep Learning (DL), have demonstrated significant potential in the interpretation of Remote Sensing (RS) imagery, covering tasks such as scene classification, object detection, land-cover/land-use classification, change detection, and multi-view stereo reconstruction. Large-scale training samples are essential for ML/DL models to achieve optimal performance. However, the current organization of training samples is ad-hoc and vendor-specific, lacking an integrated approach that can effectively manage training samples from different vendors to meet the demands of various RS AI tasks. This article proposes a solution to address these challenges by designing and implementing LuoJiaSET, a large-scale training sample database system for intelligent interpretation of RS imagery. LuoJiaSET accommodates over five million training samples, providing support for cross-dataset queries and serving as a comprehensive training data store for RS AI model training and calibration. It overcomes challenges related to label semantic categories, structural heterogeneity in label representation, and interoperable data access.

The emergence of the Digital Twin of Earth (DTE) signifies a pivotal advancement in the field of the Digital Earth, which includes observations, simulations, and predictions regarding the state of the Earth system and its temporal evolution. While the Geospatial Data Cubes (GDCs) hold promise in probing the DTE across time, existing GDC approaches are not well-suited for real-time data ingestion and processing. This paper proposes the design and implementation of a real-time ingestion and processing approach in Geospatial Data Cube (RTGDC), achieved by ingesting real-time observation streams into data cubes. The methodology employs a publish/subscribe model for efficient observation ingestion. To optimize observation processing within the data cube, a distributed streaming computing framework is utilized in the implementation of the RTGDC. This approach significantly enhances the real-time capabilities of GDCs, providing an efficient solution that brings the realization of the DTE concept within closer reach. In RTGDC, two cases involving local and global scales were implemented, and their real-time performance was evaluated based on latency, throughput, and parallel efficiency.

Classification of very high resolution imagery (VHRI) is challenging due to the difficulty in mining complex spatial and spectral patterns from rich image details. Various object-based Convolutional Neural Networks (OCNN) for VHRI classification have been proposed to overcome the drawbacks of the redundant pixel-wise CNNs, owing to their low computational cost and fine contour-preserving. However, classification performance of OCNN is still limited by geometric distortions, insufficient feature representation, and lack of contextual guidance. In this paper, an innovative multi-level context-guided classification method with the OCNN (MLCG-OCNN) is proposed. A feature-fusing OCNN, including the object contour-preserving mask strategy with the supplement of object deformation coefficient, is developed for accurate object discrimination by learning simultaneously high-level features from independent spectral patterns, geometric characteristics, and object-level contextual information. Then pixel-level contextual guidance is used to further improve the per-object classification results. The MLCG-OCNN method is intentionally tested on two validated small image datasets with limited training samples, to assess the performance in applications of land cover classification where a trade-off between time-consumption of sample training and overall accuracy needs to be found, as it is very common in the practice. Compared with traditional benchmark methods including the patch-based per-pixel CNN (PBPP), the patch-based per-object CNN (PBPO), the pixel-wise CNN with object segmentation refinement (PO), semantic segmentation U-Net (U-NET), and DeepLabV3+(DLV3+), MLCG-OCNN method achieves remarkable classification performance (> 80 %). Compared with the state-of-the-art architecture DeepLabV3+, the MLCG-OCNN method demonstrates high computational efficiency for VHRI classification (4–5 times faster).

Taxi demand forecasting plays an important role in ride-hailing services. Accurate taxi demand forecasting can assist taxi companies in pre-allocating taxis, improving vehicle utilization, reducing waiting time, and alleviating traffic congestion. It is a challenging task due to the highly non-linear and complicated spatial-temporal patterns of the taxi data. Most of the existing taxi demand forecasting methods lack the ability to capture the dynamic spatial-temporal dependencies among regions. They either fail to consider the limitations of Graph Neural Networks or do not efficiently capture the long-term temporal dependencies. In this paper, we propose a Spatial-Temporal Diffusion Convolutional Network (ST-DCN) for taxi demand forecasting. The dynamic spatial dependencies are efficiently captured through a two-phase graph diffusion convolutional network where the attention mechanism is introduced. Moreover, a novel temporal convolution module is designed to learn various ranges of temporal dependencies, including recent, daily, and weekly periods. Inside the module, convolution layers are stacked to handle very long sequences. Experimental results on two large-scale real-world taxi datasets from New York City (NYC) and Chengdu demonstrate that our method significantly outperforms seven state-of-the-art baseline methods.