As deep learning techniques are increasingly applied to gesture recognition, point cloud-based dynamic gesture recognition methods have attracted significant attention. However, many existing approaches overlook per-point motion features and the overall temporal dynamics of gestures, thereby failing to fully exploit motion and temporal cues embedded in point cloud sequences. To address this issue, we propose Motion-Temporal Network (MoTeNet), a novel framework for dynamic gesture recognition on point clouds, which preserves spatial structural information while modeling both global frame-level temporal changes and per-point motion. MoTeNet extracts spatial geometric features through a Hierarchical Graph Convolution (HGC) module, and incorporates a Motion Feature Encoding Module (MFEM) to encode point-level motion features across adjacent frames. This approach provides fine-grained dynamic information for subsequent temporal modeling. Furthermore, an Adaptive Temporal Feature Fusion (ATFF) module integrates convolutional neural networks (CNNs) and transformers to adaptively fuse short-term and long-term temporal dependencies, enabling comprehensive modeling of the dynamic evolution of gestures. Experimental results demonstrate that MoTeNet achieves state-of-the-art performance on point cloud gesture recognition benchmarks, including SHREC’17, DHG, and NVGesture, with ablation studies further validating the effectiveness of the proposed framework.

The millimeter-wave (mmWave) radar-based gait recognition technology has attracted significant attention due to its cost-effectiveness and weather resilience. The majority of existing methods are primarily designed for the recognition of gait patterns on the fixed route, and they achieve significant performance. However, fewer advancements have been made in gait pattern recognition for the free route, primarily owing to the challenges posed by data sparsity and route diversity in such contexts. To take those issues into account, we present a novel framework named Gait Spatio-Temporal Network (GSTNet), designed for efficient human gait recognition from spatiotemporal features. The GSTNet is composed of Multi-Temporal Resolution DGCNN (MTRD) and Dynamic Feature Capturing Module (DFCM). The MTDR regroups frames before edge convolution, extracting not only the spatial features of each frame but also the dynamic features from neighboring frames to cope with relative data sparsity. The DFCM adaptively generates weights for each frame through temporal features to handle diverse gait samples. Experiments demonstrate our GSTNet achieves state-of-the-art on the mmGait and STPointGCN datasets. Additionally, we provide results from the ablation study to further validate the efficacy of the proposed framework.

Visual SLAM (Simultaneous Localisation and Mapping) aims to simultaneously estimate camera poses and depth maps from navigation videos captured. While recent deep learning based methods have achieved great success on this task, they tend to work well on source domain data and suffer from performance degradation on the unseen data of target domain. Hence, we propose an online adaptation approach to continuously adapt a pre-trained visual SLAM model to changing environments in a self-supervised manner. To preserve pre-learned knowledge against catastrophic forgetting, we perform updating on a novel adapter proposed rather than fine-tuning the whole model for adaptation. The adapter includes a cross-domain feature translation module that translates pre-learned features into translated features suitable for adaptation. Ideally, the translated new features should not only contain pre-learned knowledge but also substantially distinct from pre-learned features since these two features represent different domains. We thus introduce cycle-consistent contrastive learning to maximize the dissimilarity between these two features by enlarging the distance between them in the feature space. Besides, our contrastive learning method exploiting cycle-consistency contraint enables the translated features to be transferred back to the pre-learned ones, which helps the translated features better preserve pre-learned knowledge. Comprehensive experiments on both synthetic and real-world datasets demonstrate superior adaptation performance of our proposed method over several state-of-the-art baselines.