3D Gaussian Splatting (3D-GS) technique couples 3D Gaussian primitives with differentiable rasterization to achieve high-quality novel view synthesis results while providing advanced real-time rendering performance. However, due to the flaw of its adaptive density control strategy in 3D-GS, it frequently suffers from over-reconstruction issue in intricate scenes containing high-frequency details, leading to blurry rendered images. The underlying reason for the flaw has still been under-explored. In this work, we present a comprehensive analysis of the cause of aforementioned artifacts, namely gradient collision, which prevents large Gaussians in over-reconstructed regions from splitting. To address this issue, We propose the novel homodirectional view-space positional gradient as the criterion for densification. Our strategy efficiently identifies large Gaussians in over-reconstructed regions, and recovers fine details by splitting. We evaluate our proposed method on various challenging datasets. The experimental results indicate that our approach achieves the best rendering quality with reduced or similar memory consumption. Our method is easy to implement and can be incorporated into a wide variety of most recent Gaussian Splatting-based methods. The code is publicly available at https://ty424.github.io/AbsGS.github.io

Three-dimensional convolutional neural networks (3D CNNs) have gained popularity in many complicated computer vision applications. Many customized accelerators based on FPGAs are proposed for 2D CNNs, while very few are for 3D CNNs. Three-D CNNs are far more computationally intensive and the design space for 3D CNN acceleration has been further expanded since one more dimension is introduced, making it a big challenge to accelerate 3D CNNs on FPGAs. Motivated by the finding that the computation patterns of 2D and 3D CNNs are very similar, we propose a uniform architecture design for accelerating both 2D and 3D CNNs in this paper. The uniform architecture is based on the idea of mapping convolutions to matrix multiplications. A customized mapping module is developed to generate the feature matrix tilings with no need to store the entire enlarged feature matrix on-chip or off-chip, a splitting strategy is adopted to reconstruct a convolutional layer to adapt to the on-chip memory capacity, and a 2D multiply-and-accumulate (MAC) array is adopted to compute matrix multiplications efficiently. For demonstration, we implement an accelerator prototype with a high-level synthesis (HLS) methodology on a Xilinx VC709 board and test the accelerator on three typical CNN models: AlexNet, VGG16, and C3D. Experimental results show that the accelerator achieves state-of-the-art throughput performance on both 2D and 3D CNNs, with much better energy efficiency than the CPU and GPU.

Underwater acoustic signal classification is a critical task that involves identifying different types of signals in a complex and dynamic underwater environment, which is often contaminated by strong ambient noise. Recent studies have demonstrated that deep learning-based methods can achieve remarkable performance in this task by leveraging large-scale labeled data. However, obtaining labeled data in real-world scenarios can be challenging due to the labor-intensive and expert-dependent nature of the labeling process, especially for underwater scenarios. In this study, we propose a novel self-supervised learning framework combined with mixup-based augmentation that can learn discriminative representations from large-scale unlabeled data, thereby reducing the dependence on labeled data. Additionally, we propose a Test Time Augmentation module to further improve the model's robustness. Our proposed approach achieves a classification accuracy of 86.33% on the DeepShip dataset, surpassing previous competitive methods by a significant margin. Notably, our method demonstrates excellent generalization performance in few-shot scenarios and low signal-to-noise settings, highlighting its potential for practical applications.

Hao Wang, Zhen Huang, Yong Dou, Yu Hong. Proceedings of the 28th International Conference on Computational Linguistics. 2020.

ITM Web of Conferences, open-access proceedings in information technology, computer science and mathematics