No abstract

In automatic driving, the recognition of Front-Vehicle taillights plays a key role in predicting the intentions of the vehicle ahead. In order to accurately identify the Front-Vehicle taillights, we first analyze the different characteristics of the vehicle taillight signal, and then propose an improved taillight recognition model based on YOLOv5s. First, CA(coordinate attention) is inserted into the backbone network of YOLOv5s model to improve small target recognition and reduce interference from other light sources; Then, the EIOU Loss is used to solve the class imbalance problem; Finally, EIOU-NMS is used to solve the problem of anchor box error suppression in the recognition process. We use the actual scene video and vehicle taillights dataset to conduct ablation experiments to verify the effectiveness of the improved algorithm. The experimental results show that the mAP value of the model is 9.2% higher than YOLOv5s.INDEX TERMS Autonomous driving, vehicle taillights recognition, ablation experiment.

A detection and tracking algorithm based on improved YOLOv5 is proposed for the poor recognition and tracking of obscured targets and small-sized targets. The K-means ++ algorithm is used to cluster to obtain new anchor values; the CIOU-NMS is introduced to improve the missed detection problem when the target is obscured; the CBAM is proposed to be embedded into the Backbone and Neck part to improve the feature extraction capability of the algorithm for small targets. DeepSORT is chosen as the multi-target tracker to plot the motion trajectory of the target in real time. The experimental results show that the improved algorithm has a 2.1% improvement in detection accuracy and a detection speed of 32.32/s, satisfying real-time efficient detection with better tracking.

We present VideoPoseVR, a video-based animation authoring workflow using online videos to author character animations in VR. It leverages the state-of-the-art deep learning approach to reconstruct 3D motions from online videos, caption the motions, and store them in a motion dataset. Creators can import the videos, search in the dataset, modify the motion timeline, and combine multiple motions from videos to author character animations in VR. We implemented a proof-of-concept prototype and conducted a user study to evaluate the feasibility of the video-based authoring approach as well as gather initial feedback of the prototype. The study results suggest that VideoPoseVR was easy to learn for novice users to author animations and enable rapid exploration of prototyping for applications such as entertainment, skills training, and crowd simulations.

In skeleton-based human action recognition, Transformer, which models the correlations between joint pairs in global topology, has achieved remarkable results. However, compared to many researches on changing graph topology learning in GCN, Transformer self-attention ignores the topology of the skeleton graph when capturing the dependencies between joints. To address these problems, we propose a novel two-stream spatial Graphormer network (2s-SGR), which models joint and bone information using self-attention incorporating structural encodings. First, in the joint stream, while Transformer models joint correlations in the global topology of the space, the topology of the joints and the edge information of the bones are introduced into the self-attention through custom structural encodings. At the same time, joint motion information is modeled in spatial-temporal blocks. The added information on structure and motion can effectively capture the dependencies of nodes between frames and enhance feature representation. Second, for the second-order information of the skeleton, the bone stream adapts to the structure of the bone by adjusting the custom structural encodings. Finally, the global spatial-temporal features of joints and bones in the skeleton are fused and input into the classification network to obtain action recognition results. Extensive experiments on three large-scale datasets, NTU-RGB+D 60, NTU-RGB+D 120, and Kinetics, demonstrate that the performance of the 2s-SGR proposed in this paper is at the state-of-the-art level and is effectively validated by ablation experiments.