Adapting pre-trained LiDAR segmentation models to dynamic domain shifts during testing is of paramount importance for the safety of autonomous driving. Most existing methods neglect the influence of domain changes and point density in continual test-time adaption (CTTA), relying on backpropagation and large batch sizes for stability. We approach this problem with three insights: 1) Point clouds at different distances usually have different densities resulting in distribution disparities; 2) The feature distribution of different domains varies, and domain-aware parameters can alleviate domain gaps; 3) Features are highly correlated and make segmentation of different labels confusing. To this end, this work presents D3CTTA, an online backpropagation-free framework for 3D continual test-time adaption for LiDAR segmentation. D3CTTA consists of a distance-aware prototype learning module to integrate LiDAR-based geometry prior and a domain-dependent decorrelation module to reduce feature correlations among different domains and different categories. Extensive experiments on three benchmarks showcase that our method achieves a state-of-the-art performance compared to both backpropagation-based methods and backpropagation-free methods. Code is available at https://github.com/ZhaoJichun1/D3CTTA.

This work presents a novel framework for few-shot 3D part segmentation. Recent advances have demonstrated the significant potential of 2D foundation models for low-shot 3D part segmentation. However, it is still an open problem that how to effectively aggregate 2D knowledge from foundation models to 3D. Existing methods either ignore geometric structures for 3D feature learning or neglects the high-quality grouping clues from SAM, leading to under-segmentation and inconsistent part labels. We devise a novel SAM segment graph-based propagation method, named SegGraph, to explicitly learn geometric features encoded within SAM's segmentation masks. Our method encodes geometric features by modeling mutual overlap and adjacency between segments while preserving intra-segment semantic consistency. We construct a segment graph, conceptually similar to an atlas, where nodes represent segments and edges capture their spatial relationships (overlap/adjacency). Each node adaptively modulates 2D foundation model features, which are then propagated via a graph neural network to learn global geometric structures. To enforce intra-segment semantic consistency, we map segment features to 3D points with a novel view-direction-weighted fusion attenuating contributions from low-quality segments. Extensive experiments on PartNet-E demonstrate that our method outperforms all competing baselines by at least 6.9% mIoU. Further analysis reveals that SegGraph achieves particularly strong performance on small components and part boundaries, demonstrating its superior geometric understanding.

The challenge of inverse synthetic aperture radar (ISAR) imaging under sparse aperture (SA) conditions using compressed sensing (CS) techniques has attracted significant attention. Current CS-based methods suffer from feature information degradation during iterative reconstruction, leading to sparse images lacking detailed structural characteristics. Inspired by the memory storage mechanisms of the human brain, this paper proposes a novel framework for sparse aperture ISAR imaging. The framework integrates brain-inspired cognitive proximal mapping modules (BLCPMM) within the alternating direction method of multipliers (ADMM) algorithm. To address information degradation between stages, a brain-like cognitive transitional memory (BLCTM) module is introduced. Additionally, a brain-like cognitive global memory (BLCGM) module is incorporated to ensure information continuity across stages. The proposed model, termed the Brain-Like Cognitive ADMM Network (BLCAN), effectively mitigates feature memory loss during iterative transmissions. Simulation and experimental results demonstrate that BLCAN significantly improves image quality under challenging conditions, exhibiting superior robustness and adaptability.

An increasing number of robotic programs are implemented based on Robot Operating System (ROS), which provides many practical tools and libraries for robot development. To improve robot reliability and security, several recent approaches apply fuzzing to ROS programs for bug detection. However, these approaches still have some main limitations, including inefficient test case generation, ineffective program feedback and weak generality/automation.

The sparse aperture (SA) phenomenon, when encountered in the context of inverse synthetic aperture radar (ISAR) imaging, poses a formidable challenge in acquiring high-resolution ISAR images to establish ground truth. This challenge imposes limitations on the practical implementation of data-driven SA-ISAR imaging methods. In response to this issue, we introduce a novel deep learning approach called Parallel ISTA Net (PIN) to enhance the quality of SA-ISAR imaging and reduce the reliance on high-quality labeled data. The model is an interpretable deep unfolding model that achieves self-supervised training through a parallel architecture. The PIN model uses a multi-sampling matrix training strategy to enhance the robustness of the network through the symmetry constraints provided by the parallel framework. In addition, we introduce a weighting factor to adjust the loss function to improve imaging quality further. Real-measured data imaging results show that this method can achieve robust imaging performance comparable to supervised methods at low sampling rates of 50% and 25%.

Inverse synthetic aperture radar (ISAR) imaging can be compromised by sparse apertures (SAs) leading to diminished image quality. While deep learning models exhibit significant potential for SAs ISAR imaging, existing methods often expand network capacity by stacking layers to enhance imaging outcomes, thereby complicating the training process. Furthermore, the adaptability of the network may be constrained by the fixed computational graph. To address this problem, we propose the ResDEQ model, a novel SAs ISAR imaging method based on a deep equilibrium framework and a single-layer ResNet. ResDEQ decouples the forward and backward propagation processes using fewer parameters, enhancing network flexibility and adaptability to various imaging requirements. Experimental results indicate that ResDEQ surpasses traditional layer-stacked networks in imaging performance.

Abstract Joint pose estimation of human hands and objects from a single RGB image is an important topic for AR/VR, robot manipulation, etc. It is common practice to determine both poses directly from the image; some recent methods attempt to improve the initial poses using a variety of contact‐based approaches. However, few methods take the real physical constraints conveyed by the image into consideration, leading to less realistic results than the initial estimates. To overcome this problem, we make use of a set of high‐level 2D features which can be directly extracted from the image in a new pipeline which combines contact approaches and these constraints during optimization. Our pipeline achieves better results than direct regression or contact‐based optimization: they are closer to the ground truth and provide high quality contact.

Previous online 3D dense reconstruction methods struggle to achieve the balance between memory storage and surface quality, largely due to the usage of stagnant underlying geometry representation, such as TSDF (truncated signed distance functions) or surfels, without any knowledge of the scene priors. In this paper, we present DI-Fusion (Deep Implicit Fusion), based on a novel 3D representation, i.e. Probabilistic Local Implicit Voxels (PLIVoxs), for online 3D reconstruction with a commodity RGB-D camera. Our PLIVox encodes scene priors considering both the local geometry and uncertainty parameterized by a deep neural network. With such deep priors, we are able to perform online implicit 3D reconstruction achieving state-of-the-art camera trajectory estimation accuracy and mapping quality, while achieving better storage efficiency compared with previous online 3D reconstruction approaches. Our implementation is available at https://www.github.com/huangjh-pub/di-fusion.

Abstract Reconstructing dynamic scenes with commodity depth cameras has many applications in computer graphics, computer vision, and robotics. However, due to the presence of noise and erroneous observations from data capturing devices and the inherently ill-posed nature of non-rigid registration with insufficient information, traditional approaches often produce low-quality geometry with holes, bumps, and misalignments. We propose a novel 3D dynamic reconstruction system, named HDR-Net-Fusion, which learns to simultaneously reconstruct and refine the geometry on the fly with a sparse embedded deformation graph of surfels, using a hierarchical deep reinforcement (HDR) network. The latter comprises two parts: a global HDR-Net which rapidly detects local regions with large geometric errors, and a local HDR-Net serving as a local patch refinement operator to promptly complete and enhance such regions. Training the global HDR-Net is formulated as a novel reinforcement learning problem to implicitly learn the region selection strategy with the goal of improving the overall reconstruction quality. The applicability and efficiency of our approach are demonstrated using a large-scale dynamic reconstruction dataset. Our method can reconstruct geometry with higher quality than traditional methods.