Speech Separation with Pretrained Frontend to Minimize Domain Mismatch

Speech separation seeks to separate individual speech signals from a speech mixture. Typically, most separation models are trained on synthetic data due to the unavailability of target reference in real-world cocktail party scenarios. As a result, there exists a domain gap between real and synthetic data when deploying speech separation models in real-world applications. In this paper, we propose a self-supervised domain-invariant pretrained (DIP) frontend that is exposed to mixture data without the need for target reference speech. The DIP frontend utilizes a Siamese network with two innovative pretext tasks, mixture predictive coding (MPC) and mixture invariant coding (MIC), to capture shared contextual cues between real and synthetic unlabeled mixtures. Subsequently, we freeze the DIP frontend as a feature extractor when training the downstream speech separation models on synthetic data. By pretraining the DIP frontend with the contextual cues, we expect that the speech separation skills learned from synthetic data can be effectively transferred to real data. To benefit from the DIP frontend, we introduce a novel separation pipeline to align the feature resolution of the separation models. We evaluate the speech separation quality on standard benchmarks and real-world datasets. The results confirm the superiority of our DIP frontend over existing speech separation models. This study underscores the potential of large-scale pretraining to enhance the quality and intelligibility of speech separation in real-world applications.

Enhancing Sampling Protocol for Robust Point Cloud Classification

Established sampling protocols for 3D point cloud learning, such as Farthest Point Sampling (FPS) and Fixed Sample Size (FSS), have long been recognized and utilized. However, real-world data often suffer from corrputions such as sensor noise, which violates the benignness assumption of point cloud in current protocols. Consequently, they are notably vulnerable to noise, posing significant safety risks in critical applications like autonomous driving. To address these issues, we propose an enhanced point cloud sampling protocol, PointDR, which comprises two components: 1) Downsampling for key point identification and 2) Resampling for flexible sample size. Furthermore, differentiated strategies are implemented for training and inference processes. Particularly, an isolation-rated weight considering local density is designed for the downsampling method, assisting it in performing random key points selection in the training phase and bypassing noise in the inference phase. A local-geometry-preserved upsampling is incorporated into resampling, facilitating it to maintain a stochastic sample size in the training stage and complete insufficient data in the inference. It is crucial to note that the proposed protocol is free of model architecture altering and extra learning, thus minimal efforts are demanded for its replacement of the existing one. Despite the simplicity, it substantially improves the robustness of point cloud learning, showcased by outperforming the state-of-the-art methods on multiple benchmarks of corrupted point cloud classification. The code will be available upon the paper's acceptance.

Pyramid Diffusion for Fine 3D Large Scene Generation

Directly transferring the 2D techniques to 3D scene generation is challenging due to significant resolution reduction and the scarcity of comprehensive real-world 3D scene datasets. To address these issues, our work introduces the Pyramid Discrete Diffusion model (PDD) for 3D scene generation. This novel approach employs a multi-scale model capable of progressively generating high-quality 3D scenes from coarse to fine. In this way, the PDD can generate high-quality scenes within limited resource constraints and does not require additional data sources. To the best of our knowledge, we are the first to adopt the simple but effective coarse-to-fine strategy for 3D large scene generation. Our experiments, covering both unconditional and conditional generation, have yielded impressive results, showcasing the model's effectiveness and robustness in generating realistic and detailed 3D scenes. Our code will be available to the public.

Gradient-based adaptive wavelet de-noising method for photoacoustic imaging in vivo

Photoacoustic imaging (PAI) has been applied to many biomedical applications over the past decades. However, the received PA signal usually suffers from poor signal-to-noise ratio (SNR). Conventional solution of employing higher-power laser, or doing long-time signal averaging, may raise the system cost, time consumption, and tissue damage. Another strategy is de-noising algorithm design. In this paper, we propose a new de-noising method, termed gradient-based adaptive wavelet de-noising, which sets the energy gradient mutation point of low-frequency wavelet components as the threshold. We conducted simulation, ex vivo and in vivo experiments to validate the performance of the algorithm. The quality of de-noised PA image/signal by our proposed algorithm has improved by 20%-40%, in comparison to the traditional signal denoising algorithms, which produces better contrast and clearer details. The proposed de-noising method provides potential to improve the SNR of PA signal under single-shot low-power laser illumination for biomedical applications in vivo.

Risk-optimized Outlier Removal for Robust Point Cloud Classification

The popularity of point cloud deep models for safety-critical purposes has increased, but the reliability and security of these models can be compromised by intentional or naturally occurring point cloud noise. To combat this issue, we present a novel point cloud outlier removal method called PointCVaR, which empowers standard-trained models to eliminate additional outliers and restore the data. Our approach begins by conducting attribution analysis to determine the influence of each point on the model output, which we refer to as point risk. We then optimize the process of filtering high-risk points using Conditional Value at Risk (CVaR) as the objective. The rationale for this approach is based on the observation that noise points in point clouds tend to cluster in the tail of the risk distribution, with a low frequency but a high level of risk, resulting in significant interference with classification results. Despite requiring no additional training effort, our method produces exceptional results in various removal-and-classification experiments for noisy point clouds, which are corrupted by random noise, adversarial noise, and backdoor trigger noise. Impressively, it achieves 87% accuracy in defense against the backdoor attack by removing triggers. Overall, the proposed PointCVaR effectively eliminates noise points and enhances point cloud classification, making it a promising plug-in module for various models in different scenarios.

Primitive3D: 3D Object Dataset Synthesis from Randomly Assembled Primitives

Numerous advancements in deep learning can be attributed to the access to large-scale and well-annotated datasets. However, such a dataset is prohibitively expensive in 3D computer vision due to the substantial collection cost. To alleviate this issue, we propose a cost-effective method for automatically generating a large amount of 3D objects with annotations. In particular, we synthesize objects simply by assembling multiple random primitives. These objects are thus auto-annotated with part labels originating from primitives. This allows us to perform multi-task learning by combining the supervised segmentation with unsupervised reconstruction. Considering the large overhead of learning on the generated dataset, we further propose a dataset distillation strategy to remove redundant samples regarding a target dataset. We conduct extensive experiments for the downstream tasks of 3D object classification. The results indicate that our dataset, together with multi-task pretraining on its annotations, achieves the best performance compared to other commonly used datasets. Further study suggests that our strategy can improve the model performance by pretraining and fine-tuning scheme, especially for the dataset with a small scale. In addition, pretraining with the proposed dataset distillation method can save 86\% of the pretraining time with negligible performance degradation. We expect that our attempt provides a new data-centric perspective for training 3D deep models.

PointBA: Towards Backdoor Attacks in 3D Point Cloud

3D deep learning has been increasingly more popular for a variety of tasks including many safety-critical applications. However, recently several works raise the security issues of 3D deep nets. Although most of these works consider adversarial attacks, we identify that backdoor attack is indeed a more serious threat to 3D deep learning systems but remains unexplored. We present the backdoor attacks in 3D with a unified framework that exploits the unique properties of 3D data and networks. In particular, we design two attack approaches: the poison-label attack and the clean-label attack. The first one is straightforward and effective in practice, while the second one is more sophisticated assuming there are certain data inspections. The attack algorithms are mainly motivated and developed by 1) the recent discovery of 3D adversarial samples which demonstrate the vulnerability of 3D deep nets under spatial transformations; 2) the proposed feature disentanglement technique that manipulates the feature of the data through optimization methods and its potential to embed a new task. Extensive experiments show the efficacy of the poison-label attack with over 95% success rate across several 3D datasets and models, and the ability of clean-label attack against data filtering with around 50% success rate. Our proposed backdoor attack in 3D point cloud is expected to perform as a baseline for improving the robustness of 3D deep models.

Integrated Age Estimation Mechanism

Machine-learning-based age estimation has received lots of attention. Traditional age estimation mechanism focuses estimation age error, but ignores that there is a deviation between the estimated age and real age due to disease. Pathological age estimation mechanism the author proposed before introduces age deviation to solve the above problem and improves classification capability of the estimated age significantly. However,it does not consider the age estimation error of the normal control (NC) group and results in a larger error between the estimated age and real age of NC group. Therefore, an integrated age estimation mechanism based on Decision-Level fusion of error and deviation orientation model is proposed to solve the problem.Firstly, the traditional age estimation and pathological age estimation mechanisms are weighted together.Secondly, their optimal weights are obtained by minimizing mean absolute error (MAE) between the estimated age and real age of normal people. In the experimental section, several representative age-related datasets are used for verification of the proposed method. The results show that the proposed age estimation mechanism achieves a good tradeoff effect of age estimation. It not only improves the classification ability of the estimated age, but also reduces the age estimation error of the NC group. In general, the proposed age estimation mechanism is effective. Additionally, the mechanism is a framework mechanism that can be used to construct different specific age estimation algorithms, contributing to relevant research.

Campus3D: A Photogrammetry Point Cloud Benchmark for Hierarchical Understanding of Outdoor Scene

Learning on 3D scene-based point cloud has received extensive attention as its promising application in many fields, and well-annotated and multisource datasets can catalyze the development of those data-driven approaches. To facilitate the research of this area, we present a richly-annotated 3D point cloud dataset for multiple outdoor scene understanding tasks and also an effective learning framework for its hierarchical segmentation task. The dataset was generated via the photogrammetric processing on unmanned aerial vehicle (UAV) images of the National University of Singapore (NUS) campus, and has been point-wisely annotated with both hierarchical and instance-based labels. Based on it, we formulate a hierarchical learning problem for 3D point cloud segmentation and propose a measurement evaluating consistency across various hierarchies. To solve this problem, a two-stage method including multi-task (MT) learning and hierarchical ensemble (HE) with consistency consideration is proposed. Experimental results demonstrate the superiority of the proposed method and potential advantages of our hierarchical annotations. In addition, we benchmark results of semantic and instance segmentation, which is accessible online at https://3d.dataset.site with the dataset and all source codes.