HiRo-SLAM: A High-Accuracy and Robust Visual-Inertial SLAM System with Precise Camera Projection Modeling and Adaptive Feature Selection.

HighlightsWhat are the main findings?Introducing Area-to-Point hierarchical matching into the front end can suppress cross-region mismatches in repetitive-texture scenes and improve the reliability of front-end observations.The Anchor–Explorer feature selection strategy can identify high-quality features that balance stability and geometric contribution, thereby enhancing the effectiveness of the observations used in back-end optimization.What are the implications of the main findings?In complex degraded environments with repetitive textures, low illumination, and weak textures, improving front-end observation quality and track continuity is more beneficial to visual–inertial state estimation than simply increasing the number of candidate features.The proposed method provides an extensible front-end/back-end collaborative design for visual–inertial SLAM in complex degraded environments.The localization performance of visual–inertial simultaneous localization and mapping (VI-SLAM) strongly depends on front-end feature matching. In degraded scenes with low illumination, repetitive textures, and weak textures, traditional geometric front ends often suffer from sparse features and mismatches, resulting in unstable state estimation. To address this issue, this paper proposes Area-to-Point Matching Visual–Inertial SLAM (A2PM-VINS), a visual–inertial SLAM method based on Area-to-Point matching. The method introduces Area-to-Point hierarchical matching and a kinematic temporal inheritance mechanism to improve matching reliability and track continuity, and further designs an Anchor–Explorer feature selection strategy to retain features with higher geometric value for back-end optimization. In addition, a Sub-Window Consistency (SWC) weighting strategy is incorporated into the back end to suppress geometrically deceptive observations with poor temporal continuity and geometric consistency. Experiments on the European Robotics Challenge Micro Aerial Vehicle (EuRoC MAV) dataset show that A2PM-VINS achieves superior or competitive localization accuracy on multiple challenging sequences. The absolute trajectory errors on MH_04 and MH_05 are 0.0983 m and 0.1191 m, respectively, and stable tracking is maintained on V2_02, where VINS-Fusion fails. These results show that the proposed method effectively improves the robustness of visual–inertial state estimation in complex degraded environments.

Visual inertial localization method assisted by pedestrian motion features

<div class="section abstract"><div class="htmlview paragraph">The VINS-Mono algorithm, which is based on a visual-inertial SLAM framework, faces challenges in extracting feature points in regions with weak or repetitive textures and struggles to achieve accurate localization under unstable lighting conditions. This paper proposes STO-VINS, a robust monocular visual-inertial SLAM algorithm that introduces several key innovations in feature extraction. Key innovations of STO-VINS include: (1) an adaptive multi-scale image preprocessing pipeline that combines image scaling, CLAHE enhancement, and Gaussian filtering, reducing computational complexity by 64% while maintaining feature quality; (2) bidirectional Lucas-Kanade optical flow consistency verification with geometric constraint validation, which significantly reduces false tracking rates by 30-40%; (3) a grid-based multi-feature fusion detection strategy combining Shi-Tomasi corner detection and ORB feature extraction, ensuring uniform spatial distribution of features and feature diversity; (4) an intelligent dynamic parameter adjustment system that optimizes detection parameters based on multi-dimensional image quality assessments (brightness histogram analysis, Laplacian blur metric, and Canny edge density); and (5) a smart feature point quality filtering mechanism that implements distance-based deduplication and non-maximum suppression to retain the optimal 100 feature points. These innovations offer three key advantages: enhanced robustness through multi-algorithm fusion and consistency verification, improved computational efficiency through multi-scale processing and grid-based detection, and superior environmental adaptability through intelligent parameter optimization. Experimental validation using the EuRoC dataset shows that STO-VINS achieves a 6.5% improvement in localization accuracy over VINS-Mono in non-loop closure scenarios. Further outdoor scene experiments confirm that, while VINS-Mono suffers from severe trajectory drift, STO-VINS produces trajectories that closely match the experimental route with minimal drift error. The results demonstrate that STO-VINS significantly improves feature point extraction in challenging environments and offers a new paradigm for intelligent, adaptive, and high-efficiency feature tracking in SLAM systems, leading to substantial improvements in real-time performance, system stability, and environmental adaptability.</div></div>

To mitigate the heavy reliance on semantic information and the unreliability of manual feature extraction in dynamic simultaneous localization and mapping (SLAM) and object tracking systems, a novel visual-inertial SLAM with deep feature extraction matching is proposed. A lightweight network is designed for feature extraction and description, replacing the oriented FAST and rotated BRIEF (ORB) approach, which can address deep learning latency and the shortcomings of manual extraction. A fast object tracking method is developed for extracting and associating dynamic objects based on clustering analyses of feature point pair distances to epipolar lines, which enables real-time tracking of numerous objects without semantic data. An advanced online incremental loop closure detector for deep features is designed, which supersedes ORB-based detectors and maintains global pose optimization. The system’s effectiveness and advantages, including its proficiency in real-time embedded platform execution and enhanced self-localization and dynamic object tracking, have been demonstrated through extensive evaluations. Notably, the system is capable of consistently tracking dynamic objects with rapid movement or weak texture, furnishing the localization system with robust dynamic constraints.

GCENet: A geometric correspondence estimation network for tracking and loop detection in visual-inertial SLAM

Planetary surface is a complex unstructured environment with strong light, strong shadows, weak textures and numerous obstacles. There is no GPS signal support and satellite networking support on the planetary surface. The limited computer and sensor capabilities of planetary rover have a significant gap from the ground autonomous driving configuration, which presents a great challenge to the autonomous navigation methods of planetary rover. Extreme lighting conditions will increase the camera mismatching rate, and poor texture will increase the lidar matching error, which will reduce the estimation accuracy. Moreover, camera and lidar produce great motion distortion in the rugged terrain and bumpy environment, which leads to a large increase in the cumulative error of odometer. Camera and lidar are not suitable as the main sensor of SLAM algorithm in unstructured environment because they are sensitive to the change of environment. Therefore, an autonomous navigation method for planetary rover based on multi-modal fusion and multi-factor graph optimization is presented in this paper. The IMU (Inertial Measurement Unit) odometer node is added to the lidar odometer and visual odometer thread, and the IMU is taken as the central node to build a multi-factor graph optimization model. A multi-factor graph optimization model and a strong adaptive constraint strategy with reasonable weights are constructed. The pose estimation of other sensors is used to constrain IMU bias, meanwhile historical rut tracking, obstacle contour matching and skyline feature matching are introduced. Finally, motion prediction is achieved in the IMU odometer node. Simulation results and field tests show that this method can effectively cope with the unstructured environment on planetary surface, which achieves adaptive and robust autonomous navigation of planetary rover.

This letter introduces an efficient visual-inertial simultaneous localization and mapping (SLAM) method using point and line features. Currently, point-based SLAM methods do not perform well in scenarios such as weak textures and motion blur. Many researchers have noticed the excellent properties of line features in space and have attempted to develop line-based SLAM systems. However, the vast computational effort of the line extraction and description matching process makes it challenging to guarantee the real-time performance of the whole SLAM system, and the incorrect line detection and matching limit the performance improvement of the SLAM system. In this letter, we improve the traditional line detection model by means of short-line fusion, line feature uniform distribution, and adaptive threshold extraction to obtain high-quality line features for constructing SLAM constraints. Based on the gray level invariance assumption and colinear constraint, we propose a line optical flow tracking method, which significantly improves the speed of line feature matching. In addition, a measurement model that is independent of line endpoints is presented for estimating line residuals. The experimental results show that our algorithm improves the efficiency of line feature detection and matching and localization accuracy.

BackgroundGait dynamics are widely used to diagnose neurodegenerative illnesses like ALS, Parkinson’s, and Huntington’s. Cognitive bias might hinder clinical assessments and early detection. Machine learning may help detect aberrant gait patterns early.MethodsA novel ensemble classifier, the Dynamic Weighted Prediction Framework (DWPF), and an innovative feature selection methodology, Stability-Enhanced Dynamic Thresholding (SEDT), have been proposed for neurodegenerative disease detection. DWPF combines predictions from various heterogeneous base classifiers using dynamic weighted averaging with weights based on performance. By adapting to classifier efficacies, this method improves ensemble accuracy. The DWPF’s dynamic weight allocation based on real-time classifier performance makes the model robust across datasets and classification problems. This strategy balances each base classifier's reliability to improve accuracy and model resilience to overfitting and underfitting. SEDT uses stability enhancement and dataset-specific feature relevance thresholds to identify significant neurodegenerative disease features before classification. Unlike fixed threshold approaches, SEDT may adjust to different data subsets' feature relevance, preserving only the most important features for model training. The DWPF ensemble and SEDT feature selection approach are merged to detect neurodegenerative diseases in gait data. Extensive binary classification testing has been conducted to discriminate controls from ALS, Parkinson's, and Huntington’s patients.ResultsThe proposed SEDT + DWPF model was validated using Holdout Validation, K-Fold Cross-Validation, and Monte Carlo (Shuffle-Split) Cross-Validation. Monte Carlo Cross-Validation had the best ALS detection accuracy (93.50%), sensitivity (94.63%), and specificity (92.80%). K-Fold Cross-Validation detected Huntington's disease with the best accuracy (81.09%), sensitivity (74.19%), and specificity (89.28%). Monte Carlo Cross-Validation performed best for Parkinson's disease identification, with 84.58% accuracy, 78.95% sensitivity, and 89.71% specificity.ConclusionDWPF-SEDT integration improves neurodegenerative disease diagnosis using adaptive feature selection and robust ensemble classification. This method has great potential for early and accurate neurodegenerative disease diagnosis.

The images captured under extreme lighting conditions can exhibit severe image degradation, which significantly impacts the performance of downstream visual tasks. Existing deep learning-based approaches for low-light image enhancement have primarily focused on spatial domain enhancement while neglecting frequency domain information. Therefore, we introduce a novel network for low-light image enhancement that operates in both spatial and frequency domains (multi-scale pyramid and Fourier transform) simultaneously, named SMFNet. Our main idea involves using a dual-branch structure, incorporating spatial and multi-scale frequency domain branches. The spatial branch employs SpaBlock to extract image features, while the multi-scale frequency branch utilizes Laplacian pyramid and Fourier transform to extract frequency domain information. Furthermore, SpaBlock supplements spatial domain information in the frequency branch. Extensive experiments demonstrate that the proposed approach yields promising results in terms of both quantitative and qualitative metrics across various publicly available datasets.

The rise of Next Generation Sequencing (NGS) technologies has transformed de novo genome sequencing into an accessible research tool, but obtaining high quality eukaryotic genome assemblies remains a challenge, mostly due to the abundance of repetitive elements. These also make it difficult to study nucleotide polymorphism in repetitive regions, including certain types of structural variations. One solution proposed for resolving such regions is Sequence Assembly aided by Mutagenesis (SAM), which relies on the fact that introducing enough random mutations breaks the repetitive structure, making assembly possible. Sequencing many different mutated copies permits the sequence of the repetitive region to be inferred by consensus methods. However, this approach relies on molecular cloning in order to isolate and amplify individual mutant copies, making it hard to scale-up the approach for use in conjunction with high-throughput sequencing technologies. To address this problem, we propose NG-SAM, a modified version of the SAM protocol that relies on PCR and dilution steps only, coupled to a NGS workflow. NG-SAM therefore has the potential to be scaled-up, e.g. using emerging microfluidics technologies. We built a realistic simulation pipeline to study the feasibility of NG-SAM, and our results suggest that under appropriate experimental conditions the approach might be successfully put into practice. Moreover, our simulations suggest that NG-SAM is capable of reconstructing robustly a wide range of potential target sequences of varying lengths and repetitive structures.

Weak texture and motion blur are always challenging problems for visual-inertial odometry (VIO) systems. To improve accuracy of VIO systems in the challenging scenes, we propose a point-line-based VIO system with novel feature hybrids and with novel predicting-matching for long line track. Point-line features with shorter tracks are categorized into “MSCKF” features and with longer tracks into “SLAM” features. Especially, “SLAM” lines are added into the state vector to improve accuracy of the proposed system. Besides, to ensure the reliability and stability of detection and tracking of line features, we also propose a new “Predicting-Matching” line segment tracking method to increase the track lengths of line segments. Experimental results show that the proposed method outperforms the state-of-the-art methods of VINS-Mono [1], PL-VINS [2] and OpenVINS [3]) on both a public dataset and a collected dataset in terms of accuracy. The collected dataset is full of extremely weak textures and motion blurs. On this dataset, the proposed method also obtains better accuracy than ORB-SLAM3 [4].

UMPL- VINS: Generalized SLAM for multi-scene metaverse applications

Event cameras are novel bio-inspired sensors, where individual pixels operate independently and asynchronously, generating intensity changes as events. Leveraging the microsecond resolution (no motion blur) and high dynamic range (compatible with extreme light conditions) of events, there is considerable promise in directly segmenting objects from sparse and asynchronous event streams in various applications. However, different from the rich cues in video object segmentation, it is challenging to segment complete objects from the sparse event stream. In this paper, we present the first framework for continuous-time object segmentation from event stream. Given the object mask at the initial time, our task aims to segment the complete object at any subsequent time in event streams. Specifically, our framework consists of a Recurrent Temporal Embedding Extraction (RTEE) module based on a novel ResLSTM, a Cross-time Spatiotemporal Feature Modeling (CSFM) module which is a transformer architecture with long-term and short-term matching modules, and a segmentation head. The historical events and masks (reference sets) are recurrently fed into our framework along with current-time events. The temporal embedding is updated as new events are input, enabling our framework to continuously process the event stream. To train and test our model, we construct both real-world and simulated event-based object segmentation datasets, each comprising event streams, APS images, and object annotations. Extensive experiments on our datasets demonstrate the effectiveness of the proposed recurrent architecture.

Object detection plays a crucial role in various cutting-edge applications, such as autonomous vehicles and advanced robotics systems, primarily relying on conventional frame-based RGB sensors. However, these sensors face challenges such as motion blur and poor performance under extreme lighting conditions. Novel event-based cameras, inspired by biological vision systems, offer a promising solution with superior performance in fast-motion and challenging lighting environments while consuming less power. This work explores the integration of event-based cameras with advanced object detection frameworks, introducing Recurrent YOLOv8 (ReYOLOV8), a refined object detection framework that enhances a leading frame-based YOLO detection system with spatiotemporal modeling capabilities by adding recurrency. ReYOLOv8 incorporates a low-latency, memory-efficient method for encoding event data called Volume of Ternary Event Images (VTEI) and introduces a novel data augmentation technique based on Random Polarity Suppression (RPS) optimized for event-based sensors and tailored to leverage the unique attributes of event data. The framework was evaluated using two comprehensive event-based datasets Prophesee's Generation 1 (GEN1) and Person Detection for Robotics (PEDRo). On the GEN1 dataset, ReYOLOv8 achieved mAP improvements of 5%, 2.8%, and 2.5% across nano, small, and medium scales, respectively, while reducing trainable parameters by 4.43% on average and maintaining real-time processing speeds between 9.2 ms and 15.5 ms. For the PEDRo dataset, ReYOLOv8 demonstrated mAP improvements ranging from 9% to 18%, with models reduced in size by factors of 14.5 × and 3.8 × and an average speed improvement of 1.67 × . The results demonstrate the significant potential of bio-inspired event-based vision sensors when combined with advanced object detection frameworks. In particular, the ReYOLOv8 system effectively bridges the gap between biological principles of vision and artificial intelligence, enabling robust and efficient visual processing in dynamic and complex environments. The codes are available on GitHub at the following link https://github.com/silvada95/ReYOLOv8.

Underwater robots face significant positioning challenges, as conventional terrestrial methods like GPS and infrared motion capture fail in underwater environments due to the attenuation of electromagnetic waves. In addition, acoustic-based methods suffer from signal reflections on walls in confined environments. However, centimeter-level positioning accuracy is required to perform precise inspection in environments used in industrial applications such as spent-fuel tanks in nuclear power plants or fire-protection reservoirs. Although fiducial markers offer a potential solution, inconsistent underwater lighting conditions severely reduce how reliably they can be recognized. In this study, we present an active fiducial marker system that addresses these limitations by using LED strobe illumination for reliable and accurate underwater robot positioning. The proposed system features an LED array capable of dynamically generating fiducial marker patterns using blue LEDs at a 475 nm wavelength to minimize water absorption. A vision-based synchronization strategy is integrated to enable coordination between the marker and the camera without external communication, and strobe operation is used to reduce power consumption and motion blur. The marker cell design was optimized by testing its illuminance and uniformity to achieve optimal recognition performance. The results of an experimental evaluation in an industrial water tank demonstrated centimeter-level positioning accuracy (within 2% error) and consistent detection even under extreme lighting conditions where conventional passive markers fail, while also providing vision-based communication capability by changing the marker’s ID. The system enables precise positioning for industrial water tank inspection and research with underwater robots. The proposed approach is well-suited for mobile applications and provides a cost-effective alternative to expensive underwater motion-capture systems while maintaining high positioning accuracy.