Search

Recently, 3D Gaussian Splatting (3D-GS) has achieved impressive results in novel view synthesis, demonstrating high fidelity and efficiency. However, it easily exhibits needle-like artifacts, especially when increasing the sampling rate. Mip-Splatting tries to remove these artifacts with a 3D smoothing filter for frequency constraints and a 2D Mip filter for approximated supersampling. Unfortunately, it tends to produce over-blurred results, and sometimes needle-like Gaussians still persist. Our spectral analysis of the covariance matrix during optimization and densification reveals that current 3D-GS lacks shape awareness, relying instead on spectral radius and view positional gradients to determine splitting. As a result, needle-like Gaussians with small positional gradients and low spectral entropy fail to split and overfit high-frequency details. Furthermore, both the filters used in 3D-GS and Mip-Splatting reduce the spectral entropy and increase the condition number during zooming in to synthesize novel view, causing view inconsistencies and more pronounced artifacts. Our Spectral-GS, based on spectral analysis, introduces 3D shape-aware splitting and 2D view-consistent filtering strategies, effectively addressing these issues, enhancing 3D-GS's capability to represent high-frequency details without noticeable artifacts, and achieving high-quality photorealistic rendering.

Line detection is a classic and essential problem in image processing, computer vision and machine intelligence. Line detection has many important applications, including image vectorization (e.g., document recognition and art design), indoor mapping, and important societal challenges (e.g., sea ice fracture line extraction from satellite imagery). Many line detection algorithms and methods have been developed, but robust and intuitive methods are still lacking. In this paper, we proposed and implemented a topological graph-guided algorithm, named TGGLinesPlus, for line detection. Our experiments on images from a wide range of domains have demonstrated the flexibility of our TGGLinesPlus algorithm. We also benchmarked our algorithm with five classic and state-of-the-art line detection methods and the results demonstrate the robustness of TGGLinesPlus. We hope our open-source implementation of TGGLinesPlus will inspire and pave the way for many applications where spatial science matters., Comment: Our TGGLinesPlus Python implementation is open source. 27 pages, 8 figures and 4 tables

Large language models (LLMs) garner significant attention for their unprecedented performance, leading to an increasing number of researches evaluating LLMs. However, these evaluation benchmarks are limited to assessing the instruction-following capabilities, overlooking the fundamental abilities that emerge during the pre-training stage. Previous subjective evaluation methods mainly reply on scoring by API models. However, in the absence of references, large models have shown limited ability to discern subtle differences. To bridge the gap, we propose F-Eval, a bilingual evaluation benchmark to evaluate the fundamental abilities, including expression, commonsense and logic. The tasks in F-Eval include multi-choice objective tasks, open-ended objective tasks, reference-based subjective tasks and reference-free subjective tasks. For reference-free subjective tasks, we devise new evaluation methods, serving as alternatives to scoring by API models. We conduct evaluations on 13 advanced LLMs. Results show that our evaluation methods show higher correlation coefficients and larger distinction than other evaluators. Additionally, we discuss the influence of different model sizes, dimensions, and normalization methods. We anticipate that F-Eval will facilitate the study of LLMs' fundamental abilities., Comment: ACL 2024

As the second most common neurodegenerative disease, Parkinson's disease has caused serious problems worldwide. However, the cause and mechanism of PD are not clear, and no systematic early diagnosis and treatment of PD have been established. Many patients with PD have not been diagnosed or misdiagnosed. In this paper, we proposed an EEG-based approach to diagnosing Parkinson's disease. It mapped the frequency band energy of electroencephalogram(EEG) signals to 2-dimensional images using the interpolation method and identified classification using capsule network(CapsNet) and achieved 89.34% classification accuracy for short-term EEG sections. A comparison of separate classification accuracy across different EEG bands revealed the highest accuracy in the gamma bands, suggesting that we need to pay more attention to the changes in gamma band changes in the early stages of PD., Comment: 6 pages,2 image, 3 tables

Recent development in wireless communications has provided many reliable solutions to emergency response issues, especially in scenarios with dysfunctional or congested base stations. Prior studies on underwater emergency communications, however, remain under-studied, which poses a need for combining the merits of different underwater communication links (UCLs) and the manipulability of unmanned vehicles. To realize energy-efficient underwater emergency communications, we develop a novel underwater emergency communication network (UECN) assisted by multiple links, including underwater light, acoustic, and radio frequency links, and autonomous underwater vehicles (AUVs) for collecting and transmitting underwater emergency data. First, we determine the optimal emergency response mode for an underwater sensor node (USN) using greedy search and reinforcement learning (RL), so that isolated USNs (I-USNs) can be identified. Second, according to the distribution of I-USNs, we dispatch AUVs to assist I-USNs in data transmission, i.e., jointly optimizing the locations and controls of AUVs to minimize the time for data collection and underwater movement. Finally, an adaptive clustering-based multi-objective evolutionary algorithm is proposed to jointly optimize the number of AUVs and the transmit power of I-USNs, subject to a given set of constraints on transmit power, signal-to-interference-plus-noise ratios (SINRs), outage probabilities, and energy, which achieves the best tradeoff between the maximum emergency response time (ERT) and the total energy consumption (EC). Simulation results indicate that our proposed approach outperforms benchmark schemes in terms of energy efficiency (EE), contributing to underwater emergency communications., Comment: 15 pages