MTR++: Multi-Agent Motion Prediction With Symmetric Scene Modeling and Guided Intention Querying.

Motion prediction is crucial for autonomous driving systems to understand complex driving scenarios and make informed decisions. However, this task is challenging due to the diverse behaviors of traffic participants and complex environmental contexts. In this paper, we propose Motion TRansformer (MTR) frameworks to address these challenges. The initial MTR framework utilizes a transformer encoder-decoder structure with learnable intention queries, enabling efficient and accurate prediction of future trajectories. By customizing intention queries for distinct motion modalities, MTR improves multimodal motion prediction while reducing reliance on dense goal candidates. The framework comprises two essential processes: global intention localization, identifying the agent's intent to enhance overall efficiency, and local movement refinement, adaptively refining predicted trajectories for improved accuracy. Moreover, we introduce an advanced MTR++ framework, extending the capability of MTR to simultaneously predict multimodal motion for multiple agents. MTR++ incorporates symmetric context modeling and mutually-guided intention querying modules to facilitate future behavior interaction among multiple agents, resulting in scene-compliant future trajectories. Extensive experimental results demonstrate that the MTR framework achieves state-of-the-art performance on the highly-competitive motion prediction benchmarks, while the MTR++ framework surpasses its precursor, exhibiting enhanced performance and efficiency in predicting accurate multimodal future trajectories for multiple agents.

ST3D++: Denoised Self-Training for Unsupervised Domain Adaptation on 3D Object Detection.

In this paper, we present a self-training method, named ST3D++, with a holistic pseudo label denoising pipeline for unsupervised domain adaptation on 3D object detection. ST3D++ aims at reducing noise in pseudo label generation as well as alleviating the negative impacts of noisy pseudo labels on model training. First, ST3D++ pre-trains the 3D object detector on the labeled source domain with random object scaling (ROS) which is designed to reduce target domain pseudo label noise arising from object scale bias of the source domain. Then, the detector is progressively improved through alternating between generating pseudo labels and training the object detector with pseudo-labeled target domain data. Here, we equip the pseudo label generation process with a hybrid quality-aware triplet memory to improve the quality and stability of generated pseudo labels. Meanwhile, in the model training stage, we propose a source data assisted training strategy and a curriculum data augmentation policy to effectively rectify noisy gradient directions and avoid model over-fitting to noisy pseudo labeled data. These specific designs enable the detector to be trained on meticulously refined pseudo labeled target data with denoised training signals, and thus effectively facilitate adapting an object detector to a target domain without requiring annotations. Finally, our method is assessed on four 3D benchmark datasets (i.e., Waymo, KITTI, Lyft, and nuScenes) for three common categories (i.e., car, pedestrian and bicycle). ST3D++ achieves state-of-the-art performance on all evaluated settings, outperforming the corresponding baseline by a large margin (e.g., 9.6%  âˆ¼  38.16% on Waymo → KITTI in terms of AP[Formula: see text]), and even surpasses the fully supervised oracle results on the KITTI 3D object detection benchmark with target prior. Code is available at https://github.com/CVMI-Lab/ST3D.

From Points to Parts: 3D Object Detection From Point Cloud With Part-Aware and Part-Aggregation Network.

3D object detection from LiDAR point cloud is a challenging problem in 3D scene understanding and has many practical applications. In this paper, we extend our preliminary work PointRCNN to a novel and strong point-cloud-based 3D object detection framework, the part-aware and aggregation neural network (Part- A2 net). The whole framework consists of the part-aware stage and the part-aggregation stage. First, the part-aware stage for the first time fully utilizes free-of-charge part supervisions derived from 3D ground-truth boxes to simultaneously predict high quality 3D proposals and accurate intra-object part locations. The predicted intra-object part locations within the same proposal are grouped by our new-designed RoI-aware point cloud pooling module, which results in an effective representation to encode the geometry-specific features of each 3D proposal. Then the part-aggregation stage learns to re-score the box and refine the box location by exploring the spatial relationship of the pooled intra-object part locations. Extensive experiments are conducted to demonstrate the performance improvements from each component of our proposed framework. Our Part- A2 net outperforms all existing 3D detection methods and achieves new state-of-the-art on KITTI 3D object detection dataset by utilizing only the LiDAR point cloud data.

Applicability of Yielding-Resisting Sand Column and Three-Dimensional Coordination Support in Stopes.

In view of the existing problems of stope roadways, which are difficult to maintain under the influence of high ground and mining-induced stresses, the structural characteristics and movement regularities of stopes surrounding rocks were analysed. Through the construction of a three-dimensional mechanical model of the coordination support of a stope, the adaptability index of the support in stope is presented, and its mechanism of operation is expounded. Yielding-resisting sand column (YRSC) sidewall-support technology with satisfactory compressibility and supporting strength was developed. The structure and actual mechanical properties of the YRSC were investigated through laboratory experiments, and the optimum ratio of filling materials was obtained. The good applicability of the load and deformation adaptability index of the three-dimensional coordination support in the stope and YRSC sidewall-support technology were demonstrated in practice at the No. 12306 working face of the Dongda coal mine. It was shown that the designed carrying capacity and compression of the sand columns satisfied the site requirements. The actual stress and deformation of the YRSC exhibited three stages: Slow growth at the initial stage, a large increase in the medium term, and a stable trend at the end. The adaptability index of the three-dimensional coordination support in the stope considers all bearing structure units of the stope as an interconnected whole, and the stability conditions of the stope roadway can be quantitatively described. The supporting effect of the YRSC is remarkable and can be applied to the construction of tunnels, bridge systems and other engineering fields.