Abstract We propose LeafFit, a pipeline that converts 3D Gaussian Splatting (3DGS) of individual plants into editable, instanced mesh assets. While 3DGS faithfully captures complex foliage, its high memory footprint and lack of mesh topology make it incompatible with traditional game production workflows. We address this by leveraging the repetition of leaf shapes; our method segments leaves from the unstructured 3DGS, with optional user interaction included as a fallback. A representative leaf group is selected and converted into a thin, sharp mesh to serve as a template; this template is then fitted to all other leaves via differentiable Moving Least Squares (MLS) deformation. At runtime, the deformation is evaluated efficiently on‐the‐fly using a vertex shader to minimize storage requirements. Experiments demonstrate that LeafFit achieves higher segmentation quality and deformation accuracy than recent baselines while significantly reducing data size and enabling parameter‐level editing. Our source code is publicly available at https://github.com/netbeifeng/leaf_fit .

We introduce Motion2VecSets, a 4D diffusion model for dynamic surface mesh generation from various ambiguous observations, including a sequence of RGB images, sparse and partial point clouds, and low-resolution voxel grids. While recent methods using neural field representations have shown success in modeling non-rigid objects, conventional feed-forward architectures struggle with noisy, partial, or sparse observations due to their deterministic nature. To address the inherent one-to-many mapping problem, we introduce a diffusion model that explicitly learns the shape and motion distribution of non-rigid objects through an iterative denoising process of compressed latent representations. The diffusion-based priors provide more plausible and diverse reconstructions under ambiguous conditions. Instead of relying on global latent codes, we represent 4D dynamics using latent sets. This novel 4D representation captures local shape and deformation patterns, leading to more accurate non-linear motion capture and significantly improving generalization capacity to unseen motions and identities. For temporally coherent tracking, we jointly denoise latent sets across frames and enable cross-frame information exchange. To reduce computational cost, we design an interleaved spatial-temporal attention block that alternately aggregates deformation latents along spatial and temporal dimensions. Extensive experiments on datasets of humans, animals, and articulated objects demonstrate that Motion2VecSets outperforms prior methods in reconstructing and tracking non-rigid deformations from various imperfect observations.

We introduce Motion2VecSets, a 4D diffusion model for dynamic surface reconstruction from point cloud sequences. While existing state-of-the-art methods have demonstrated success in reconstructing non-rigid objects using neural field representations, conventional feed-forward networks encounter challenges with ambiguous observations from noisy, partial, or sparse point clouds. To address these challenges, we introduce a diffusion model that explicitly learns the shape and motion distribution of non-rigid objects through an iterative denoising process of compressed latent representations. The diffusion-based priors enable more plausible and probabilistic reconstructions when handling ambiguous inputs. We parameterize 4D dynamics with latent sets instead of using global latent codes. This novel 4D representation allows us to learn local shape and deformation patterns, leading to more accurate nonlinear motion capture and significantly improving generalizability to unseen motions and identities. For more temporally-coherent object tracking, we synchronously denoise deformation latent sets and exchange information across multiple frames. To avoid computational overhead, we designed an interleaved space and time attention block to alternately aggregate deformation latents along spatial and temporal domains. Extensive comparisons against state-of-the-art methods demonstrate the superiority of our Motion2VecSets in 4D reconstruction from various imperfect observations.

In deep learning (DL)-based hyperspectral imagery classification, “spatial patching” is primarily used as a preprocessing for incorporating local spatial information. This operation can help to promote classification accuracy but it is facing new challenges in the unmanned aerial vehicle (UAV)-borne hyperspectral imagery with high spatial and spectral resolutions (H2imagery). The ground objects’ various spatial scales result in it being challenging to determine the optimal size for the spatial patches. In addition, due to the severe spectral variability and spatial heterogeneity of the H2imagery, “spatial patching” only exploits the local spatial information and results in serious salt-and-pepper (SP) noise and isolated areas in the classification maps. In this article, to address these issues, a novel spectral patching network (SPNet) with an end-to-end DL architecture is proposed for UAV-borne H2imagery classification. The “spectral patching” approach is proposed to preserve the global spatial information and almost all the spectral information of the original hyperspectral imagery. An end-to-end deep encoder–decoder network is then constructed based on the spectral patching mechanism, which introduces the deep residual network (ResNet) and atrous spatial pyramid pooling (ASPP) modules to extract multiscale high-level semantic information for the H2imagery classification. The experimental results obtained with the Wuhan UAV-borne H2imagery (WHU-Hi) UAV-borne hyperspectral data set demonstrate that SPNet can achieve state-of-the-art accuracy and visualization performance in the classification of H2imagery.

Large scale crop mapping is an important task in agricultural resource monitoring. To obtain a distribution map of crops, traditional methods usually require the well-designed manufacture features and the ground-truth labels of all land-cover types for training a multi-class classifier. However, the redundant labeling for each land-cover type is time-consuming and labor-intensive, and the feature design for different remote sensing data is complex and limited to human prior knowledge. In this paper, a deep one-class crop extraction framework is proposed to solve the problems mentioned above. Specifically, it uses the deep one-class crop extraction module to extract the feature automatically for any remote sensing imagery and the one-class crop extraction loss to address the lack of negative class in the deep one-class classification model. In addition, the proposed framework can be applied to multi-modal remote sensing data, i.e. hyperspectral, multispectral, and SAR images, which is verified in the experiments and the proposed framework can achieve the highest accuracy on each multi-modal data.

The Deep Learning (DL) methods can automatically extract features without artificial prior, provides an effective solution for multi-temporal crop classification. Convolutional Neural Networks (CNNs) have superb spatial-spectral feature extraction capabilities, but often lack consideration of the temporal relationship of multi-temporal images, while Recurrent Neural Networks (RNNs) can better learn the sequential variation pattern. This paper designed a deep spatial-temporal-spectral feature learning network (CropNet) combining the advantages of a deep spatial-spectral feature learning module and a deep temporal-spectral feature learning module for better feature extraction in crop classification from time-series remote sensing images. From the results, the proposed method has better crop classification effects from time-series multispectral images in our experimental areas compared with some common traditional machine learning approaches and common DL methods.

Deep learning (DL)-based hyperspectral classification primarily use "spatial patching" as preprocessing for incorporating local spatial information. This operation can help to promote classification accuracy but faces the following problems. First, it is difficult to determine the optimal size of spatial patches for different hyperspectral images (HSIs). Second, this operation only exploits spatial features locally but not globally. In this paper, we propose a novel spectral patching network (SPNet) with an end-to-end deep learning architecture for HSI classification. SPNet uses "spectral patching" and Atrous Spatial Pyramid Pooling (ASPP) module to fully preserve the local and global spatial contextual information of original HSIs. The experimental results with UAV-borne hyperspectral dataset demonstrate that the SPNet achieved state-of-the-art accuracy and visualization performance in.

The effects of the number of design variables on the optimization performances in Kriging-model-based manyobjective optimizations, which use expected hypervolume improvement (EHVI), expected improvement (EI), and estimation (EST) as the criteria for updating the Kriging model, are investigated based on four independent performance metrics in this paper. Numerical experiments are conducted in 3 to 15-objective DTLZ1 and DTLZ7 problems. The results indicate that the advantages of EHVI over EI and EST are more obvious when the number of design variables increases, and EHVI is more suitable for the problems with a large number of design variables. In addition, the comparison results show that, EHVI obtains faster IGD reduction than EI and EST in most test problems. The advantage of EHVI over EI and EST is mainly shown on the convergence performance. The spread performance is better in both EHVI and EI considering estimation errors than EST without considering estimation errors. However, the uniformity of EHVI is weak, especially for the problems with a large number of objectives.

The many-objective optimization performance of using expected hypervolume improvement (EHVI) as the updating criterion of the Kriging surrogate model is investigated, and compared with those of using expected improvement (EI) and estimation (EST) updating criteria in this paper. An exact algorithm to calculate hypervolume is used for the problems with less than six objectives. On the other hand, in order to improve the efficiency of hypervolume calculation, an approximate algorithm to calculate hypervolume based on Monte Carlo sampling is adopted for the problems with more objectives. Numerical experiments are conducted in 3 to 12-objective DTLZ1, DTLZ2, DTLZ3 and DTLZ4 problems. The results show that, in DTLZ3 problem, EHVI always obtains better convergence and diversity performances than EI and EST for any number of objectives. In DTLZ2 and DTLZ4 problems, the advantage of EHVI is shown gradually as the number of objectives increases. The present results suggest that EHVI will be a highly competitive updating criterion for the many-objective optimization with the Kriging model.

By providing an integrated and optimized support for user-defined aggregates (UDAs), data stream management systems (DSMS) can achieve superior power and generality while preserving compatibility with current SQL standards. This is demonstrated by the Stream Mill system that, through is Expressive Stream Language (ESL), efficiently supports a wide range of applications - including very advanced ones such as data stream mining, streaming XML processing, time-series queries, and RFID event processing. ESL supports physical and logical windows (with optional slides and tumbles) on both built-in aggregates and UDAs, using a simple framework that applies uniformly to both aggregate functions written in an external procedural languages and those natively written in ESL. The constructs introduced in ESL extend the power and generality of DSMS, and are conducive to UDA-specific optimization and efficient execution as demonstrated by several experiments.

ESL1 enables users to develop stream applications in an SQL-like high level language that provides the ease-of-use of a declarative language, which is Turing complete in terms of expressive power [11].

An ever-increasing amount of valuable information is stored in Web databases, "hidden" behind search interfaces. To save the user's effort in manually exploring each database, metasearchers automatically select the most relevant databases to a user's query. In this paper, we focus on one of the technical challenges in metasearching, namely database selection. Past research uses a precollected summary of each database to estimate its "relevancy" to the query, and in many cases make incorrect database selection. In this paper, we propose two techniques: probabilistic relevancy modelling and adaptive probing. First, we model the relevancy of each database to a given query as a probabilistic distribution, derived by sampling that database. Using the probabilistic model, the user can explicitly specify a desired level of certainty for database selection. The adaptive probing technique decides which and how many databases to contact in order to satisfy the user's requirement. Our experiments on real hidden-Web databases indicate that our approach significantly improves the accuracy of database selection at the cost of a small number of database probing.