Temporal Feature Fusion for 3D Detection in Monocular Video.

Previous monocular 3D detection works focus on the single frame input in both training and inference. In real-world applications, temporal and motion information naturally exists in monocular video. It is valuable for 3D detection but under-explored in monocular works. In this paper, we propose a straightforward and effective method for temporal feature fusion, which exhibits low computation cost and excellent transferability, making it conveniently applicable to various monocular models. Specifically, with the help of optical flow, we transform the backbone features produced by prior frames and fuse them into the current frame. We introduce the scene feature propagating mechanism, which accumulates history scene features without extra time-consuming. In this process, occluded areas are removed via forward-backward scene consistency. Our method naturally introduces valuable temporal features, facilitating 3D reasoning in monocular 3D detection. Furthermore, accumulated history scene features via scene propagating mitigate heavy computation overheads for video processing. Experiments are conducted on variant baselines, which demonstrate that the proposed method is model-agonistic and can bring significant improvement to multiple types of single-frame methods.

CrossFormer++: A Versatile Vision Transformer Hinging on Cross-Scale Attention.

While features of different scales are perceptually important to visual inputs, existing vision transformers do not yet take advantage of them explicitly. To this end, we first propose a cross-scale vision transformer, CrossFormer. It introduces a cross-scale embedding layer (CEL) and a long-short distance attention (LSDA). On the one hand, CEL blends each token with multiple patches of different scales, providing the self-attention module itself with cross-scale features. On the other hand, LSDA splits the self-attention module into a short-distance one and a long-distance counterpart, which not only reduces the computational burden but also keeps both small-scale and large-scale features in the tokens. Moreover, through experiments on CrossFormer, we observe another two issues that affect vision transformers' performance, i.e., the enlarging self-attention maps and amplitude explosion. Thus, we further propose a progressive group size (PGS) paradigm and an amplitude cooling layer (ACL) to alleviate the two issues, respectively. The CrossFormer incorporating with PGS and ACL is called CrossFormer++. Extensive experiments show that CrossFormer++ outperforms the other vision transformers on image classification, object detection, instance segmentation, and semantic segmentation tasks.

Finite element analysis for mechanics of guiding catheters in transfemoral intervention.

BACKGROUND: During the percutaneous coronary intervention (PCI), the strong backup support of a guiding catheters is essential in reaching a target coronary lesion successfully. Nevertheless, it is difficult to explore the mechanics of a guiding catheter by analytical and experimental methods due to its complex deformation and interactions among guiding catheter, guide wire, and artery. In this study, the finite element method was used to analyze the backup support of a guiding catheter in transfemoral intervention (TFI). METHODS: A finite element model was established in the light of geometric, mechanical properties of the guiding catheter and boundary con ditions. To validate the finite element model, an arterial tree model was constructed to measure the backup force of the guiding catheters in TFI. Then, the process of the guiding catheter disengaged from the ostium was analyzed. RESULTS: The influencing rules of the geometric parameters of the guiding catheter on its backup support in TFI were obtained with the help of the finite element model. The result shows that the larger the outer diameter and wall thickness, the greater the backup support, 2.0 to 2.3 mm of the outer diameter was suggested. When designing a guiding catheter, it is wise to avoid α, the angle between the line and wall of the artery, being within 60° and 75° and it is a better choice if the contact length is between 5 and 12 mm. CONCLUSIONS: This study sought to investigate the influencing rules of the geometric parameters of the guiding catheter on its backup support in transfemoral intervention. A finite element model for analyzing the backup support of a guiding catheter was validated by experiments. It indicated that the finite element method can analyze the varied laws of the guiding catheter with different geometric parameters.