People spend approximately 70% of their time indoors. Understanding the indoor environments is therefore important for a wide range emerging mobile personal and social applications. Knowledge floorplans often required by these However, are either unavailable or obtaining them requires slow, tedious, error-prone manual labor.

Deep neural networks have demonstrated remarkable reconstruction for single-image super-resolution (SISR). However, most existing CNN-based SISR methods directly learn the relation between low-resolution (LR) and high-resolution (HR) images, neglecting to explore recurrence of internal patches, hence hindering representational power CNNs. In this paper, we propose a novel single image Super-Resolution network based on Graph ATtention (SRGAT) make full use patch-recurrence in natural image....

Deep subspace clustering network has shown its effectiveness in hyperspectral image (HSI) clustering. However, there are two major challenges that need to be addressed: 1) lack of effective supervision for feature learning; and 2) negative effect caused by the high redundancy global dictionary atoms. In this article, we propose an end-to-end trainable HSI Specifically, ensure extracted features well-suited subsequent clustering, cluster assignments with confidence employed as pseudo-labels...

The purpose of non-uniformity and blind pixel correction is to provide a more reliable foundation for subsequent image processing target detection. Existing methods generally struggle balance the contradiction between over-smoothing residual noise. Particularly, can easily filter out texture details dim small targets. Based on multi-frame response model infrared focal plane array detector, we propose two-stage 3-D fully convolutional network factor estimation, integrated with an suppression...

Acid fracturing fluids can effectively improve the microporous structure of coal, thereby enhancing permeability coal seam and efficiency gas drainage. To explore effects acid on pore modification samples from different ranks, hydrochloric acid-based were prepared used to soak four types medium high-rank in an experiment. High-pressure mercury intrusion liquid nitrogen adsorption techniques results demonstrated that fluid alter coal. However, effect does not exhibit a linear relationship...

The escalating demand for long-context applications has intensified the necessity of extending LLM context windows. Despite recent fine-tuning approaches successfully expanding lengths, their high memory footprints, especially activations, present a critical practical limitation. Current parameter-efficient methods prioritize reducing parameter update overhead over addressing activation constraints. Similarly, existing sparsity mechanisms improve computational efficiency but overlook...

Hyperspectral anomaly detection (HAD) aims to distinguish anomalies from background-by-background modeling. Deep learning has been applied HAD and achieves promising results. However, there exist several issues that need be addressed: 1) unrealistic Gaussian assumption on the latent representations may limit its application; 2) deep features are not well-suited due separation between feature detection; 3) lack of adequate exploitation spectral-spatial features; 4) negative effect caused by...

Hyperspectral anomaly detection (HAD) aims at distinguishing anomalies from background in an unsupervised manner. Autoencoder (AE) and its variant-based methods have achieved promising performance HAD. However, most existing neglect to exploit the local structure information of hyperspectral images (HSIs) that reflects underlying relationships between each pixel surroundings. Hence, representation capabilities networks are restricted. Moreover, reconstruction during training compels learn...

Recently, state-of-the-art classification performance of natural images has been obtained by self-supervised learning (S2L) as it can generate latent features through between different views the same images. However, semantic information similar hardly exploited these S2L-based methods. Consequently, to explore potential S2L samples in hyperspectral image (HSIC), we propose nearest neighboring (N2SSL) method, interacting augmentations reliable pairs (RN2Ps) HSI framework bootstrap your own...

Mobile devices are becoming a primary medium for personal information gathering, management, and sharing. Managing image data on mobile platforms is difficult problem due to large set size, content diversity, heterogeneous individual usage patterns, resource constraints. This article presents user-centric system, called iScope, management sharing devices. iScope uses multi-modality clustering of both context efficient search, online learning techniques predicting images interest. It also...

With great significance in military and civilian applications, subpixel target detection is of interest hyperspectral remote sensing. The targets usually also need to be unmixed identify their components. Traditionally, these are first detected then obtain corresponding abundances. Therefore, unmixing independently performed. However, there potential relations between two processes that investigated. In this article, we integrate using iterative constrained sparse representation. main idea...

This paper presents our 2nd place solution for the NuPlan Challenge 2023. Autonomous driving in real-world scenarios is highly complex and uncertain. Achieving safe planning multimodal a challenging task. Our approach, Imitation with Spatial-Temporal Heatmap, adopts learning form of behavior cloning, innovatively predicts future states heatmap representation, uses trajectory refinement techniques to ensure final safety. The experiment shows that method effectively balances vehicle's progress...

Three-dimensional motion estimation from multiview video sequences is of vital importance to achieve high-quality dynamic scene reconstruction. In this paper, we propose a new 3-D method based on matrix completion. Taking reconstructed mesh as the underlying representation, automatically estimates motions objects. A “separating + merging” framework introduced estimation. separating step, initial are first estimated for each view with neighboring view. Then, in merging obtained by merged...

We consider the analysis of an AIDS dataset where each patient is characterized by a list symptoms and labeled with one or more TCM syndromes. The task to build classifier that maps use minimum reference set-based multiple instance learning (MRS-MIL) method. method identifies representative for syndrome builds Gaussian mixture model based on them. models all syndromes are then used classification via Bayes rule. By relying subset key classification, MRS-MIL can produce reliable high quality...

Clustering of hyperspectral images is a fundamental but challenging task. The recent development image clustering has evolved from shallow models to deep and achieved promising results in many benchmark datasets. However, their poor scalability, robustness, generalization ability, mainly resulting offline scenarios, greatly limit application large-scale data. To circumvent these problems, we present scalable online model, named Spectral-Spatial Contrastive (SSCC), based on self-supervised...

The soil particle size distribution (PSD) is a fundamental physical property that can affect nutrients, structure characterization and hydraulic properties. However, the effect of plant expansion on PSD not clear. Therefore, in this study, fractal theory was applied to quantitatively describe PSD. On Mount Tai, artificial afforestation with Robinia pseudoacacia helpful restoring ecosystems improving quality. clonal spread allows R. easily escape cultivation, leading formation mixed forests....

Temporal-dense 3-D reconstruction for dynamic scenes is a challenging and important research topic in signal processing. Although can be captured by multiple high frame rate cameras, price, large storage are still problematic practical applications. To address this problem, we propose new method temporal-densely capturing reconstructing with low which consists of spatio-temporal sampling, interpolation, fusion. In fusion, dual-tree discrete wavelet transform shape context employed to compute...