Exploring Point-BEV Fusion for 3D Point Cloud Object Tracking with Transformer.

With the prevalent use of LiDAR sensors in autonomous driving, 3D point cloud object tracking has received increasing attention. In a point cloud sequence, 3D object tracking aims to predict the location and orientation of an object in consecutive frames. Motivated by the success of transformers, we propose Point Tracking TRansformer (PTTR), which efficiently predicts high-quality 3D tracking results in a coarse-to-fine manner with the help of transformer operations. PTTR consists of three novel designs. 1) Instead of random sampling, we design Relation-Aware Sampling to preserve relevant points to the given template during subsampling. 2) We propose a Point Relation Transformer for effective feature aggregation and feature matching between the template and search region. 3) Based on the coarse tracking results, we employ a novel Prediction Refinement Module to obtain the final refined prediction through local feature pooling. In addition, motivated by the favorable properties of the Bird's-Eye View (BEV) of point clouds in capturing object motion, we further design a more advanced framework named PTTR++, which incorporates both the point-wise view and BEV representation to exploit their complementary effect in generating high-quality tracking results. PTTR++ substantially boosts the tracking performance on top of PTTR with low computational overhead. Extensive experiments over multiple datasets show that our proposed approaches achieve superior 3D tracking accuracy and efficiency. Code will be available at https://github.com/Jasonkks/PTTR.

Meta-DETR: Image-Level Few-Shot Detection With Inter-Class Correlation Exploitation.

Few-shot object detection has been extensively investigated by incorporating meta-learning into region-based detection frameworks. Despite its success, the said paradigm is still constrained by several factors, such as (i) low-quality region proposals for novel classes and (ii) negligence of the inter-class correlation among different classes. Such limitations hinder the generalization of base-class knowledge for the detection of novel-class objects. In this work, we design Meta-DETR, which (i) is the first image-level few-shot detector, and (ii) introduces a novel inter-class correlational meta-learning strategy to capture and leverage the correlation among different classes for robust and accurate few-shot object detection. Meta-DETR works entirely at image level without any region proposals, which circumvents the constraint of inaccurate proposals in prevalent few-shot detection frameworks. In addition, the introduced correlational meta-learning enables Meta-DETR to simultaneously attend to multiple support classes within a single feedforward, which allows to capture the inter-class correlation among different classes, thus significantly reducing the misclassification over similar classes and enhancing knowledge generalization to novel classes. Experiments over multiple few-shot object detection benchmarks show that the proposed Meta-DETR outperforms state-of-the-art methods by large margins. The implementation codes are publicly available at https://github.com/ZhangGongjie/Meta-DETR.

Trichoderma longibrachiatum-Associated Skin Inflammation and Atypical Hyperplasia in Mouse.

Background: The relationship between infection and tumors has attracted increasing attention. Trichoderma spp. are often isolated from tumors. However, their potential role remains unclear. We recently reported the isolation of Trichoderma longibrachiatum from a patient with pulmonary spindle cell carcinoma that was confirmed as primary infection by application of laser capture microdissection and polymerase chain reaction. To explore whether the strain is pathogenic and whether it can cause atypical cell proliferation and infiltration of NK cells and T cells, we designed a mouse infection experiment. Methods: Twelve ICR mice were randomly separated into three groups. Cyclophosphamide was used to inhibit the immunity of mice. A mouse model of Trichoderma infection was successfully established by intracutaneous injection on the back skin with a suspension of strain PKUT180420015. The pathological manifestations of Trichoderma infection and the interaction between immune cells and fungi were observed by histopathology, immunohistochemistry and intensive fungal staining. Reisolation of the fungus was observed by infected tissue culture. The inoculated sites exhibited swelling 3 days after inoculation, and ulcers developed at approximately 14 days. Skin specimens were obtained and then cultured at 3, 7, and 14 days after inoculation. We selected mice 14 days after inoculation in Group 3, whose ulcers were the most typical, for histological analysis. Results: Inflammation, angioinvasion and necrosis were observed. Immunohistochemistry showed positive markers of Ki67, CD3, CD56, GZMB, and PRF. Periodic acid-Schiff staining, periodic acid-silver methenamine staining, and Calcofluor staining showed fungal spores in the vascular lumen, vascular walls and around the blood vessels. Conclusions: Our studies showed that a T. longibrachiatum strain (PKUT180420015) isolated from a biopsy specimen in a patient with pulmonary spindle cell carcinoma could induce atypical hyperplasia, with the expression of Ki67, CD3, CD56, GZMB, and PRF in mice. These data indicate that the fungus may be involved in inducing atypical hyperplasia or tumorigenesis.

[Sustained negative pressure drainage for treatment of severe maxillofacial and neck space infection].

OBJECTIVE: This study observed the curative effect of sustained negative pressure drainage application on treatment of severe maxillofacial and neck space infection. METHODS: Incision and drainage were performed to treat 18 patients with severe maxillofacial and neck space infection. A small incision was made on the site of the most obvious swelling or fluctuations, and localized negative pressure was applied with a drainage device on the wound during suturing. RESULTS: Among the 18 patients, 14 were healed, whereas 4 underwent dehiscence of the wound after the operation. Negative pressure was lost as the drainage tubes were removed, and non-negative pressure drainage method was used instead. During the negative pressure treatment, swelling and pain did not increase after the operation. Other complications, such as asphyxia, septic shock, or mediastinal abscess, did not occur. All the patients were healed and eventually discharged from the hospital. CONCLUSION: Sustained negative pressure drainage, which is a modified version of the traditional method of incision and drainage, is an alternative treatment for severe maxillofacial and neck space infection. Such treatment reduces patient pain and eases doctor exertion. Thus, this method provides a new therapeutic strategy for severe maxillofacial and neck space infection.