The Contrastive Language-Image Pretraining (CLIP) model has been widely used in various downstream vision tasks. The few-shot learning paradigm has been widely adopted to augment its capacity for these tasks. However, current paradigms may struggle with fine-grained classification, such as satellite image recognition, due to widening domain gaps. To address this limitation, we propose retrieval-enhanced visual prompt learning (RePrompt), which introduces retrieval mechanisms to cache and reuse the knowledge of downstream tasks. RePrompt constructs a retrieval database from either training examples or external data if available, and uses a retrieval mechanism to enhance multiple stages of a simple prompt learning baseline, thus narrowing the domain gap. During inference, our enhanced model can reference similar samples brought by retrieval to make more accurate predictions. A detailed analysis reveals that retrieval helps to improve the distribution of late features, thus, improving generalization for downstream tasks. RePrompt attains state-of-the-art performance on a wide range of vision datasets, including 11 image datasets, 3 video datasets, 1 multi-view dataset, and 4 domain generalization benchmarks.

The emergence of 3D Gaussian Splatting (3DGS) has recently ignited a renewed wave of research in dense visual SLAM. However, existing approaches encounter challenges, including sensitivity to artifacts and noise, suboptimal selection of training viewpoints, and the absence of global optimization. In this paper, we propose GSORB-SLAM, a dense SLAM framework that integrates 3DGS with ORB features through a tightly coupled optimization pipeline. To mitigate the effects of noise and artifacts, we propose a novel geometric representation and optimization method for tracking, which significantly enhances localization accuracy and robustness. For high-fidelity mapping, we develop an adaptive Gaussian expansion and regularization method that facilitates compact yet expressive scene modeling while suppressing redundant primitives. Furthermore, we design a hybrid graph-based viewpoint selection mechanism that effectively reduces overfitting and accelerates convergence. Extensive evaluations across various datasets demonstrate that our system achieves state-of-the-art performance in both tracking precision and rendering quality.

Summary: This paper proposes a dynamic window with virtual goal (DW-VG) method for local collision avoidance in dynamic environments. Firstly, the debounce filter and polynomial curve-fitting algorithm are combined to predict the trajectory of the obstacles with timestamps. Based on the motion prediction of the obstacles, the virtual goal is proposed to replace the real goal, so that the robot can escape from the concave trap and avoid the dynamic obstacles. According to the timestamps and virtual goal, the optimal linear and angular velocities are selected from the dynamic window, which drive the robot toward its real goal. The simulation and experimental results show that the DW-VG method can not only escape the local minima and avoid dynamic obstacles but also is applicable to the dense environment. Furthermore, the simulation results also verify that the DW-VG method drives the robot to reach its goal faster and safer than other reactive obstacle avoidance methods.

This paper investigates the issue of adaptive optimal tracking control for nonlinear systems with dynamic state constraints. An asymmetric time-varying int