Yihong Wu

Bingxi Liu, Yujie Fu, Feng Lu et al.

Visual Place Recognition (VPR) is the task of retrieving database images similar to a query photo by comparing it to a large database of known images. In real-world applications, extreme illumination changes caused by query images taken at night pose a significant obstacle that VPR needs to overcome. However, a training set with day-night correspondence for city-scale, street-level VPR does not exist. To address this challenge, we propose a novel pipeline that divides VPR and conquers Nocturnal Place Recognition (NPR). Specifically, we first established a street-level day-night dataset, NightStreet, and used it to train an unpaired image-to-image translation model. Then we used this model to process existing large-scale VPR datasets to generate the VPR-Night datasets and demonstrated how to combine them with two popular VPR pipelines. Finally, we proposed a divide-and-conquer VPR framework and provided explanations at the theoretical, experimental, and application levels. Under our framework, previous methods can significantly improve performance on two public datasets, including the top-ranked method.

Yuwen Heng, Yihong Wu, Jiawen Chen et al.

Achieving accurate material segmentation for 3-channel RGB images is challenging due to the considerable variation in a material's appearance. Hyperspectral images, which are sets of spectral measurements sampled at multiple wavelengths, theoretically offer distinct information for material identification, as variations in intensity of electromagnetic radiation reflected by a surface depend on the material composition of a scene. However, existing hyperspectral datasets are impoverished regarding the number of images and material categories for the dense material segmentation task, and collecting and annotating hyperspectral images with a spectral camera is prohibitively expensive. To address this, we propose a new model, the MatSpectNet to segment materials with recovered hyperspectral images from RGB images. The network leverages the principles of colour perception in modern cameras to constrain the reconstructed hyperspectral images and employs the domain adaptation method to generalise the hyperspectral reconstruction capability from a spectral recovery dataset to material segmentation datasets. The reconstructed hyperspectral images are further filtered using learned response curves and enhanced with human perception. The performance of MatSpectNet is evaluated on the LMD dataset as well as the OpenSurfaces dataset. Our experiments demonstrate that MatSpectNet attains a 1.60% increase in average pixel accuracy and a 3.42% improvement in mean class accuracy compared with the most recent publication. The project code is attached to the supplementary material and will be published on GitHub.

Bingxi Liu, Pengju Zhang, Li He et al.

Recent visual place recognition (VPR) approaches have leveraged foundation models (FM) and introduced novel aggregation techniques. However, these methods have failed to fully exploit key concepts of FM, such as the effective utilization of extensive training sets, and they have overlooked the potential of classical aggregation methods, such as GeM and NetVLAD. Building on these insights, we revive classical feature aggregation methods and develop more fundamental VPR models, collectively termed SuperPlace. First, we introduce a supervised label alignment method that enables training across various VPR datasets within a unified framework. Second, we propose G$^2$M, a compact feature aggregation method utilizing two GeMs, where one GeM learns the principal components of feature maps along the channel dimension and calibrates the output of the other. Third, we propose the secondary fine-tuning (FT$^2$) strategy for NetVLAD-Linear (NVL). NetVLAD first learns feature vectors in a high-dimensional space and then compresses them into a lower-dimensional space via a single linear layer. Extensive experiments highlight our contributions and demonstrate the superiority of SuperPlace. Specifically, G$^2$M achieves promising results with only one-tenth of the feature dimensions compared to recent methods. Moreover, NVL-FT$^2$ ranks first on the MSLS leaderboard.

Muzhi Li, Jinhu Qi, Yihong Wu et al.

Retrieval-augmented generation agents development is hindered by the lack of process-level supervision to effectively guide agentic capabilities like task decomposition, retriever invocation, and stepwise decision-making. While reinforcement learning offers a potential solution, it suffers from sparse rewards and the limited reasoning capabilities of large language models (LLMs). Meanwhile, existing data synthesis methods only produce chain-of-thought rationales and fail to model environmental interactions. In this paper, we propose EviPath, an evidence-anchored reasoning path synthesis paradigm for RAG agent development. EviPath comprises: (i) Abductive Subtask Planning, which decomposes the problem into sub-questions and iteratively plans an optimal solution path based on the dependencies between them; (ii) Faithful Sub-question Answering, which uses supporting evidence to construct a proxy environment to generate reasoning thoughts and answers for each sub-question; and (iii) Conversational Fine-Tuning, which formats the complete agent-environment interaction trajectory into a dialogue format suitable for Supervised Fine-Tuning. EviPath allows LLMs to learn complex reasoning and tool-use capabilities directly from synthesized data. Extensive experiments on widely-used question-answering benchmarks show that an 8B parameter model trained with EviPath-synthesized data significantly and consistently outperforms state-of-the-art baselines with a double-digit absolute EM gain of 14.7% in open-domain question answering.

Chade Li, Pengju Zhang, Yihong Wu

The rapid advancement of vision-language models (VLMs) in 3D domains has accelerated research in text-query-guided point cloud processing, though existing methods underperform in point-level segmentation due to inadequate 3D-text alignment that limits local feature-text context linking. To address this limitation, we propose MR-COSMO, a Visual-Text Memory Recall and Direct CrOSs-MOdal Alignment Method for Query-Driven 3D Segmentation, establishing explicit alignment between 3D point clouds and text/2D image data through a dedicated direct cross-modal alignment module while implementing a visual-text memory module with specialized feature banks. This direct alignment mechanism enables precise fusion of geometric and semantic features, while the memory module employs specialized banks storing text features, visual features, and their correspondence mappings to dynamically enhance scene-specific representations via attention-based knowledge recall. Comprehensive experiments across 3D instruction, reference, and semantic segmentation benchmarks confirm state-of-the-art performance.

Zewen Xu, Yijia He, Hao Wei et al.

Most existing visual-inertial odometry (VIO) initialization methods rely on accurate pre-calibrated extrinsic parameters. However, during long-term use, irreversible structural deformation caused by temperature changes, mechanical squeezing, etc. will cause changes in extrinsic parameters, especially in the rotational part. Existing initialization methods that simultaneously estimate extrinsic parameters suffer from poor robustness, low precision, and long initialization latency due to the need for sufficient translational motion. To address these problems, we propose a novel VIO initialization method, which jointly considers extrinsic orientation and gyroscope bias within the normal epipolar constraints, achieving higher precision and better robustness without delayed rotational calibration. First, a rotation-only constraint is designed for extrinsic orientation and gyroscope bias estimation, which tightly couples gyroscope measurements and visual observations and can be solved in pure-rotation cases. Second, we propose a weighting strategy together with a failure detection strategy to enhance the precision and robustness of the estimator. Finally, we leverage Maximum A Posteriori to refine the results before enough translation parallax comes. Extensive experiments have demonstrated that our method outperforms the state-of-the-art methods in both accuracy and robustness while maintaining competitive efficiency.

Chade Li, Pengju Zhang, Jiaming Zhang et al.

3D point cloud segmentation has a wide range of applications in areas such as autonomous driving, augmented reality, virtual reality and digital twins. The point cloud data collected in real scenes often contain small objects and categories with small sample sizes, which are difficult to handle by existing networks. In this regard, we propose a point cloud segmentation network that fuses local attention based on density perception with global attention. The core idea is to increase the effective receptive field of each point while reducing the loss of information about small objects in dense areas. Specifically, we divide different sized windows for local areas with different densities to compute attention within the window. Furthermore, we consider each local area as an independent token for the global attention of the entire input. A category-response loss is also proposed to balance the processing of different categories and sizes of objects. In particular, we set up an additional fully connected layer in the middle of the network for prediction of the presence of object categories, and construct a binary cross-entropy loss to respond to the presence of categories in the scene. In experiments, our method achieves competitive results in semantic segmentation and part segmentation tasks on several publicly available datasets. Experiments on point cloud data obtained from complex real-world scenes filled with tiny objects also validate the strong segmentation capability of our method for small objects as well as small sample categories.

Fan Wang, Chaofan Zhang, Fulin Tang et al.

Mapping and localization are two essential tasks for mobile robots in real-world applications. However, largescale and dynamic scenes challenge the accuracy and robustness of most current mature solutions. This situation becomes even worse when computational resources are limited. In this paper, we present a novel lightweight object-level mapping and localization method with high accuracy and robustness. Different from previous methods, our method does not need a prior constructed precise geometric map, which greatly releases the storage burden, especially for large-scale navigation. We use object-level features with both semantic and geometric information to model landmarks in the environment. Particularly, a learning topological primitive is first proposed to efficiently obtain and organize the object-level landmarks. On the basis of this, we use a robot-centric mapping framework to represent the environment as a semantic topology graph and relax the burden of maintaining global consistency at the same time. Besides, a hierarchical memory management mechanism is introduced to improve the efficiency of online mapping with limited computational resources. Based on the proposed map, the robust localization is achieved by constructing a novel local semantic scene graph descriptor, and performing multi-constraint graph matching to compare scene similarity. Finally, we test our method on a low-cost embedded platform to demonstrate its advantages. Experimental results on a large scale and multi-session real-world environment show that the proposed method outperforms the state of arts in terms of lightweight and robustness.

Yujie Fu, Yihong Wu

Most image matching methods perform poorly when encountering large scale changes in images. To solve this problem, firstly, we propose a scale-difference-aware image matching method (SDAIM) that reduces image scale differences before local feature extraction, via resizing both images of an image pair according to an estimated scale ratio. Secondly, in order to accurately estimate the scale ratio, we propose a covisibility-attention-reinforced matching module (CVARM) and then design a novel neural network, termed as Scale-Net, based on CVARM. The proposed CVARM can lay more stress on covisible areas within the image pair and suppress the distraction from those areas visible in only one image. Quantitative and qualitative experiments confirm that the proposed Scale-Net has higher scale ratio estimation accuracy and much better generalization ability compared with all the existing scale ratio estimation methods. Further experiments on image matching and relative pose estimation tasks demonstrate that our SDAIM and Scale-Net are able to greatly boost the performance of representative local features and state-of-the-art local feature matching methods.

Pengju Zhang, Yihong Wu, Jiagang Zhu

Non-local operations are usually used to capture long-range dependencies via aggregating global context to each position recently. However, most of the methods cannot preserve object shapes since they only focus on feature similarity but ignore proximity between central and other positions for capturing long-range dependencies, while shape-awareness is beneficial to many computer vision tasks. In this paper, we propose a Semi-Global Shape-aware Network (SGSNet) considering both feature similarity and proximity for preserving object shapes when modeling long-range dependencies. A hierarchical way is taken to aggregate global context. In the first level, each position in the whole feature map only aggregates contextual information in vertical and horizontal directions according to both similarity and proximity. And then the result is input into the second level to do the same operations. By this hierarchical way, each central position gains supports from all other positions, and the combination of similarity and proximity makes each position gain supports mostly from the same semantic object. Moreover, we also propose a linear time algorithm for the aggregation of contextual information, where each of rows and columns in the feature map is treated as a binary tree to reduce similarity computation cost. Experiments on semantic segmentation and image retrieval show that adding SGSNet to existing networks gains solid improvements on both accuracy and efficiency.

Pengju Zhang, Yihong Wu, Bingxi Liu

Visual localization to compute 6DoF camera pose from a given image has wide applications such as in robotics, virtual reality, augmented reality, etc. Two kinds of descriptors are important for the visual localization. One is global descriptors that extract the whole feature from each image. The other is local descriptors that extract the local feature from each image patch usually enclosing a key point. More and more methods of the visual localization have two stages: at first to perform image retrieval by global descriptors and then from the retrieval feedback to make 2D-3D point correspondences by local descriptors. The two stages are in serial for most of the methods. This simple combination has not achieved superiority of fusing local and global descriptors. The 3D points obtained from the retrieval feedback are as the nearest neighbor candidates of the 2D image points only by global descriptors. Each of the 2D image points is also called a query local feature when performing the 2D-3D point correspondences. In this paper, we propose a novel parallel search framework, which leverages advantages of both local and global descriptors to get nearest neighbor candidates of a query local feature. Specifically, besides using deep learning based global descriptors, we also utilize local descriptors to construct random tree structures for obtaining nearest neighbor candidates of the query local feature. We propose a new probabilistic model and a new deep learning based local descriptor when constructing the random trees. A weighted Hamming regularization term to keep discriminativeness after binarization is given in the loss function for the proposed local descriptor. The loss function co-trains both real and binary descriptors of which the results are integrated into the random trees.

Fulin Tang, Yihong Wu

Camera pose tracking attracts much interest both from academic and industrial communities, of which the methods based on planar markers are easy to be implemented. However, most of the existing methods need to identify multiple points in the marker images for matching to space points. Then, PnP and RANSAC are used to compute the camera pose. If cameras move fast or are far away from markers, matching is easy to generate errors and even RANSAC cannot remove incorrect matching. Then, the result by PnP cannot have good performance. To solve this problem, we design circular markers and represent 6D camera pose analytically and unifiedly as very concise forms from each of the marker by projective invariance. Afterwards, the pose is further optimized by a proposed nonlinear cost function based on a polar-n-direction geometric distance. The method is from imaged circle edges and without PnP/RANSAC, making pose tracking robust and accurate. Experimental results show that the proposed method outperforms the state of the arts in terms of noise, blur, and distance from camera to marker. Simultaneously, it can still run at about 100 FPS on a consumer computer with only CPU.