Open-vocabulary 3D visual grounding and reasoning aim to localize objects in a scene based on implicit language descriptions, even when they are occluded. This ability is crucial for tasks such as vision-language navigation and autonomous robotics. However, current methods struggle because they rely heavily on fine-tuning with 3D annotations and mask proposals, which limits their ability to handle diverse semantics and common knowledge required for effective reasoning. In this work, we propose ReasonGrounder, an LVLM-guided framework that uses hierarchical 3D feature Gaussian fields for adaptive grouping based on physical scale, enabling open-vocabulary 3D grounding and reasoning. ReasonGrounder interprets implicit instructions using large vision-language models (LVLM) and localizes occluded objects through 3D Gaussian splatting. By incorporating 2D segmentation masks from the SAM and multi-view CLIP embeddings, ReasonGrounder selects Gaussian groups based on object scale, enabling accurate localization through both explicit and implicit language understanding, even in novel, occluded views. We also contribute ReasoningGD, a new dataset containing over 10K scenes and 2 million annotations for evaluating open-vocabulary 3D grounding and amodal perception under occlusion. Experiments show that ReasonGrounder significantly improves 3D grounding accuracy in real-world scenarios.

Visual imitation learning is effective for robots to learn versatile tasks. However, many existing methods rely on behavior cloning with supervised historical trajectories, limiting their 3D spatial and 4D spatiotemporal awareness. Consequently, these methods struggle to capture the 3D structures and 4D spatiotemporal relationships necessary for real-world deployment. In this work, we propose 4D Diffusion Policy (DP4), a novel visual imitation learning method that incorporates spatiotemporal awareness into diffusion-based policies. Unlike traditional approaches that rely on trajectory cloning, DP4 leverages a dynamic Gaussian world model to guide the learning of 3D spatial and 4D spatiotemporal perceptions from interactive environments. Our method constructs the current 3D scene from a single-view RGB-D observation and predicts the future 3D scene, optimizing trajectory generation by explicitly modeling both spatial and temporal dependencies. Extensive experiments across 17 simulation tasks with 173 variants and 3 real-world robotic tasks demonstrate that the 4D Diffusion Policy (DP4) outperforms baseline methods, improving the average simulation task success rate by 16.4% (Adroit), 14% (DexArt), and 6.45% (RLBench), and the average real-world robotic task success rate by 8.6%.

Accurate 6D pose estimation is key for robotic manipulation, enabling precise object localization for tasks like grasping. We present RAG-6DPose, a retrieval-augmented approach that leverages 3D CAD models as a knowledge base by integrating both visual and geometric cues. Our RAG-6DPose roughly contains three stages: 1) Building a Multi-Modal CAD Knowledge Base by extracting 2D visual features from multi-view CAD rendered images and also attaching 3D points; 2) Retrieving relevant CAD features from the knowledge base based on the current query image via our ReSPC module; and 3) Incorporating retrieved CAD information to refine pose predictions via retrieval-augmented decoding. Experimental results on standard benchmarks and real-world robotic tasks demonstrate the effectiveness and robustness of our approach, particularly in handling occlusions and novel viewpoints. Supplementary material is available on our project website: https://sressers.github.io/RAG-6DPose.

Open-vocabulary 3D visual grounding aims to localize target objects based on free-form language queries, which is crucial for embodied AI applications such as autonomous navigation, robotics, and augmented reality. Learning 3D language fields through neural representations enables accurate understanding of 3D scenes from limited viewpoints and facilitates the localization of target objects in complex environments. However, existing language field methods struggle to accurately localize instances using spatial relations in language queries, such as ''the book on the chair.'' This limitation mainly arises from inadequate reasoning about spatial relations in both language queries and 3D scenes. In this work, we propose SpatialReasoner, a novel neural representation-based framework with large language model (LLM)-driven spatial reasoning that constructs a visual properties-enhanced hierarchical feature field for open-vocabulary 3D visual grounding. To enable spatial reasoning in language queries, SpatialReasoner fine-tunes an LLM to capture spatial relations and explicitly infer instructions for the target, anchor, and spatial relation. To enable spatial reasoning in 3D scenes, SpatialReasoner incorporates visual properties (opacity and color) to construct a hierarchical feature field. This field represents language and instance features using distilled CLIP features and masks extracted via the Segment Anything Model (SAM). The field is then queried using the inferred instructions in a hierarchical manner to localize the target 3D instance based on the spatial relation in the language query. Notably, SpatialReasoner is not limited to a specific 3D neural representation; it serves as a framework adaptable to various representations, such as Neural Radiance Fields (NeRF) or 3D Gaussian Splatting (3DGS). Extensive experiments show that our framework can be seamlessly integrated into different neural representations, outperforming baseline models in 3D visual grounding while empowering their spatial reasoning capability. Project Homepage:ZhenyangLiu.github.io/SpatialReasoner.

An ultrawideband multiple-input-multiple-output (MIMO) circularly polarized (CP) cube antenna is proposed for wireless communication and sensing of Internet of Things. This antenna comprises an ultrawideband grounded criss-cross radiator, four square parasitic patches, and a four-port feeding network. Following the inverted-F radiation principle, a criss-cross patch is incorporated with its center grounded. Surrounding it, four small exciting patches are strategically positioned to drive via electromagnetic couplings. The ultrawideband radiation characteristics are analyzed based on the mode behaviors predicted by characteristic mode analysis (CMA). The introduction of four square parasitic patches between the radiator and the ground plane leads to an improvement in impedance matching. To create a single-port CP antenna element, a four-port feeding network with equal magnitude, but spaced 90° apart in phase, is connected to the radiator. The MIMO diversity performance is assessed through metrics like envelope correlation coefficient (ECC), diversity gain (DG), mean effective gain (MEG), and channel capacity loss (CCL). To validate the design concept, a prototype of the cube antenna is fabricated and assembled, demonstrating the ability to radiate right-hand circularly polarized (RHCP) waves. The measured results indicate impressive performance metrics: a −10-dB impedance bandwidth of 2.2–7.8 GHz (112%), a 3-dB axial ratio bandwidth of 2.1–7.3 GHz (110.6%), a realized gain ranging from 5.6 to 9.5 dBic, and a radiation efficiency between 65% and 88.5%. Additionally, the port isolation exceeds 30 dB, and the ECCs are below 0.009.

The recognition of vehicle manufacturer logo is a crucial and very challenging problem, which is still an area with few published effective methods. This paper proposes a new fast and reliable system for Vehicle Logo Recognition (VLR) based on Bag-of-Words (BoW). In our system, vehicle logo images are represented as histograms of visual words and classified by SVM in three steps: firstly, extract dense-SIFT features; secondly, quantize features into visual words by `Soft-assignment' thirdly, build histograms of visual words with spatial information. Compared with traditional VLR methods, experiment results show that our proposed system achieves higher recognition accuracy with less processing time. The proposed system is evaluated on a dataset of 840 low-resolution vehicle logo images with about 30×30 pixels, which verifies that our system is practical and effective.

In the research of image quality assessment (IQA), no-reference approaches are usually thought of as a big challenge since none of original image information is available. To tackle this problem, we propose a new no-reference image quality metric through combining two recently proposed reduced-reference IQA models, namely the free energy based distortion metric (FEDM) and the structural degradation model (SDM). In this work, it will be shown that there exists an approximate linear relationship between the original image information of the free energy feature and the structural degradation information. Based on this observation and the application of support vector machine (SVM) that is widely used in the current study of IQA, our newly developed No-reference Free energy and Structural degradation based Distortion Metric (NFSDM) is found to alleviate the dependance of original images, and has achieved remarkably well prediction accuracy, outperforming the most two full-reference IQA approaches PSNR/SSIM and several mainstream no-reference image quality metrics.