Integrating 3D Gaussian Splatting (3DGS) for dense scene reconstruction has recently gained significant attention in the field of Visual Simultaneous Localization and Mapping (V-SLAM). However, the static scene assumption underlying both V-SLAM and 3DGS limits their effectiveness in real-world environments populated with dynamic objects. Dynamic objects not only degrade SLAM tracking performance but also compromise the spatial-temporal consistency of the reconstructed map, leading to severe system failures. In this work, we propose DGS-SLAM, a novel 3DGS-based V-SLAM system capable of robust self-localization and dense mapping in dynamic environments. To address dynamic scenes, DGS-SLAM integrates several key strategies: 1) object association that fuses visual and geometric information to match objects between adjacent frames; 2) motion check that extends object association to a long-term sliding window to accurately perceive the movement of objects; 3) local fine-tuning that repairs the 3DGS model and updates keyframe poses after eliminating dynamic objects; 4) keyframe selection that promptly selects static keyframes to optimize the static regions in the 3DGS model. Extensive experiments on the TUM RGB-D and BONN RGB-D Dynamic datasets demonstrate that DGS-SLAM significantly improves localization accuracy in dynamic scenes while generating high-quality static maps, compared to other existing state-of-the-art 3DGS-based methods.

Federated Learning (FL) is an emerging distributed machine learning technology. However, in practical applications, it frequently encounters the challenge of client resource heterogeneity. This can result in long wait times or even model training failures during the communication process of FL. To address this problem, we propose a dynamic hierarchical federated learning mechanism for financial models (DHFM-FLM). The local client adopts a dynamic model training design that leverages the property of resource heterogeneity to enhance the performance of the local model. To avoid prolonged wait times for failing clients, a dynamic communication detection design is proposed at three critical junctures. In each round, the model hierarchical reservation communication design is employed to collect models in segments, thus reducing communication congestion and preventing malicious attacks on the communication process. Experiments with heterogeneous computation and communication resources demonstrate that utilizing the DHFM-FLM boosts model performance by approximately 5-8% and reduces communication time by about 15%. Additionally, DHFM-FLM increases the success rate of the FL task by approximately 6%.

Recent research in inductive reasoning has focused on predicting missing links between entities that are not observed during training. However, most approaches usually require that the relations are known at the inference time. In the real world, new entities and new relations usually emerge concurrently, which greatly challenges the model's generalization ability. In this paper, we propose a novel inductive knowledge graph embedding model that effectively handles unknown entities and relations by capturing their local structural features. Specifically, a relation graph is constructed to learn relation representations. In the relation graph, we employ a four-dimensional vector to represent the interaction patterns between nodes (relations), where each dimension corresponds to a specific type of interaction. For entity representations, our model dynamically initializes entity features using relation features and attentively aggregates neighboring features of entities to update entity features. By modeling interaction patterns between relations and incorporating structural information of entities, our model learns how to aggregate neighboring embeddings using attention mechanisms, thus generating high-quality embeddings for new entities and relations. Extensive experiments on benchmark datasets demonstrate that our model outperforms state-of-the-art methods, particularly in scenarios involving completely new relations.

Abstract The high‐level semantic information extracted from the pedestrian attribute feature is an important element for pedestrian recognition. Pedestrian attribute recognition plays an important role in both intelligent video surveillance and pedestrian re‐identification promoting the convenience of searching and performance of model. This paper tries finding a practical method to improve the performance of the pedestrian re‐identification by combining pedestrian attributes and identities. The multi‐task learning method combines pedestrian recognition and attribute information in a direct way that considers the correlation between pedestrian attributes and identities but ignores the principle and degree of such correlation. To solve this problem, a new pedestrian recognition framework based on attribute mining and reasoning is proposed in this paper. To enhance the expression ability of attribute features, it designs spatial channel attention module (SCAM) based on attention mechanism to extract features from every attribute. SCAM can not only locate the attributes on the feature map, but also effectively mine channel features with a higher degree of association with attributes. In addition, both spatial attention model and channel attention model are integrated by multiple groups of parallel branches, which further improve the network performance. Finally, using the semantic reasoning and information transmission function of graph convolutional network, the relationship between attribute features and pedestrian features can be mined. Besides, pedestrian features with stronger expression ability can also be obtained. Experiment work is conducted in two databases, DukeMTMC‐reID and Market‐1501, which are commonly used in pedestrian recognition tasks. On the Market‐1501 dataset, the final effect of the algorithm model CMC‐1 can reach 94.74%, and mAP can reach 87.02%; on the DukeMTMC‐reID dataset, CMC‐1 can reach 87.03%, and mAP can reach 77.11%. The results show that our method is at the top of the existing pedestrian recognition methods.

Person Re‐identification (Re‐ID) is aimed at solving the matching problem of the same pedestrian at a different time and in different places. Due to the cross‐device condition, the appearance of different pedestrians may have a high degree of similarity; at this time, using the global features of pedestrians to match often cannot achieve good results. In order to solve these problems, we designed a Spatial Attention Network Guided by Attribute Label (SAN‐GAL), which is a dual‐trace network containing both attribute classification and Re‐ID. Different from the previous approach of simply adding a branch of attribute binary classification network, our SAN‐GAL is mainly divided into two connecting steps. First, with attribute labels as guidance, we generate Attribute Attention Heat map (AAH) through Grad‐CAM algorithm to accurately locate fine‐grained attribute areas of pedestrians. Then, the Attribute Spatial Attention Module (ASAM) is constructed according to the AHH which is taken as the prior knowledge and introduced into the Re‐ID network to assist in the discrimination of the Re‐ID task. In particular, our SAN‐GAL network can integrate the local attribute information and global ID information of pedestrians without introducing additional attribute region annotation, which has good flexibility and adaptability. The test results on Market1501 and DukeMTMC‐reID show that our SAN‐GAL can achieve good results and can achieve 85.8% Rank‐1 accuracy on DukeMTMC‐reID dataset, which is obviously competitive compared with most Re‐ID algorithms.