Search

Unsupervised 3D object detection aims to identify objects of interest from unlabeled raw data, such as LiDAR points. Recent approaches usually adopt pseudo 3D bounding boxes (3D bboxes) from clustering algorithm to initialize the model training, and then iteratively updating both pseudo labels and the trained model. However, pseudo bboxes inevitably contain noises, and such inaccurate annotation accumulates to the final model, compromising the performance. Therefore, in an attempt to mitigate the negative impact of pseudo bboxes, we introduce a new uncertainty-aware framework. In particular, Our method consists of two primary components: uncertainty estimation and uncertainty regularization. (1) In the uncertainty estimation phase, we incorporate an extra auxiliary detection branch alongside the primary detector. The prediction disparity between the primary and auxiliary detectors is leveraged to estimate uncertainty at the box coordinate level, including position, shape, orientation. (2) Based on the assessed uncertainty, we regularize the model training via adaptively adjusting every 3D bboxes coordinates. For pseudo bbox coordinates with high uncertainty, we assign a relatively low loss weight. Experiment verifies that the proposed method is robust against the noisy pseudo bboxes, yielding substantial improvements on nuScenes and Lyft compared to existing techniques, with increases of 6.9% in AP$_{BEV}$ and 2.5% in AP$_{3D}$ on nuScenes, and 2.2% in AP$_{BEV}$ and 1.0% in AP$_{3D}$ on Lyft., Comment: Preprint, 14 pages, 4 figures, 4 tables

The unsupervised 3D object detection is to accurately detect objects in unstructured environments with no explicit supervisory signals. This task, given sparse LiDAR point clouds, often results in compromised performance for detecting distant or small objects due to the inherent sparsity and limited spatial resolution. In this paper, we are among the early attempts to integrate LiDAR data with 2D images for unsupervised 3D detection and introduce a new method, dubbed LiDAR-2D Self-paced Learning (LiSe). We argue that RGB images serve as a valuable complement to LiDAR data, offering precise 2D localization cues, particularly when scarce LiDAR points are available for certain objects. Considering the unique characteristics of both modalities, our framework devises a self-paced learning pipeline that incorporates adaptive sampling and weak model aggregation strategies. The adaptive sampling strategy dynamically tunes the distribution of pseudo labels during training, countering the tendency of models to overfit easily detected samples, such as nearby and large-sized objects. By doing so, it ensures a balanced learning trajectory across varying object scales and distances. The weak model aggregation component consolidates the strengths of models trained under different pseudo label distributions, culminating in a robust and powerful final model. Experimental evaluations validate the efficacy of our proposed LiSe method, manifesting significant improvements of +7.1% AP$_{BEV}$ and +3.4% AP$_{3D}$ on nuScenes, and +8.3% AP$_{BEV}$ and +7.4% AP$_{3D}$ on Lyft compared to existing techniques., Comment: Accepted by ECCV'24, 18 pages, 5 figures, 6 tables

Large AI models, or foundation models, are models recently emerging with massive scales both parameter-wise and data-wise, the magnitudes of which can reach beyond billions. Once pretrained, large AI models demonstrate impressive performance in various downstream tasks. A prime example is ChatGPT, whose capability has compelled people's imagination about the far-reaching influence that large AI models can have and their potential to transform different domains of our lives. In health informatics, the advent of large AI models has brought new paradigms for the design of methodologies. The scale of multi-modal data in the biomedical and health domain has been ever-expanding especially since the community embraced the era of deep learning, which provides the ground to develop, validate, and advance large AI models for breakthroughs in health-related areas. This article presents a comprehensive review of large AI models, from background to their applications. We identify seven key sectors in which large AI models are applicable and might have substantial influence, including 1) bioinformatics; 2) medical diagnosis; 3) medical imaging; 4) medical informatics; 5) medical education; 6) public health; and 7) medical robotics. We examine their challenges, followed by a critical discussion about potential future directions and pitfalls of large AI models in transforming the field of health informatics., Comment: This article has been accepted for publication in IEEE Journal of Biomedical and Health Informatics

This paper introduces an innovative physics-informed deep learning framework for metamodeling of nonlinear structural systems with scarce data. The basic concept is to incorporate physics knowledge (e.g., laws of physics, scientific principles) into deep long short-term memory (LSTM) networks, which boosts the learning within a feasible solution space. The physics constraints are embedded in the loss function to enforce the model training which can accurately capture latent system nonlinearity even with very limited available training datasets. Specifically for dynamic structures, physical laws of equation of motion, state dependency and hysteretic constitutive relationship are considered to construct the physics loss. In particular, two physics-informed multi-LSTM network architectures are proposed for structural metamodeling. The satisfactory performance of the proposed framework is successfully demonstrated through two illustrative examples (e.g., nonlinear structures subjected to ground motion excitation). It turns out that the embedded physics can alleviate overfitting issues, reduce the need of big training datasets, and improve the robustness of the trained model for more reliable prediction. As a result, the physics-informed deep learning paradigm outperforms classical non-physics-guided data-driven neural networks., Comment: 21 pages, 13 figures

Seismic events, among many other natural hazards, reduce due functionality and exacerbate vulnerability of in-service buildings. Accurate modeling and prediction of building's response subjected to earthquakes makes possible to evaluate building performance. To this end, we leverage the recent advances in deep learning and develop a physics-guided convolutional neural network (PhyCNN) framework for data-driven seismic response modeling and serviceability assessment of buildings. The proposed PhyCNN approach is capable of accurately predicting building's seismic response in a data-driven fashion without the need of a physics-based analytical/numerical model. The basic concept is to train a deep PhyCNN model based on available seismic input-output datasets (e.g., from simulation or sensing) and physics constraints. The trained PhyCNN can then used as a surrogate model for structural seismic response prediction. Available physics (e.g., the law of dynamics) can provide constraints to the network outputs, alleviate overfitting issues, reduce the need of big training datasets, and thus improve the robustness of the trained model for more reliable prediction. The trained surrogate model is then utilized for fragility analysis given certain limit state criteria (e.g., the serviceability state). In addition, an unsupervised learning algorithm based on K-means clustering is also proposed to partition the limited number of datasets to training, validation and prediction categories, so as to maximize the use of limited datasets. The performance of the proposed approach is demonstrated through three case studies including both numerical and experimental examples. Convincing results illustrate that the proposed PhyCNN paradigm outperforms conventional pure data-based neural networks., Comment: 24 pages, 17 figures