Traffic sign detection and recognition using fully convolutional network guided proposals

Weakly supervised segmentation methods using bounding box annotations focus on obtaining a pixel-level mask from each box containing an object. Existing methods typically depend on a class-agnostic mask generator, which operates on the low-level information intrinsic to an image. In this work, we utilize higher-level information from the behavior of a trained object detector, by seeking the smallest areas of the image from which the object detector produces almost the same result as it does from the whole image. These areas constitute a bounding-box attribution map (BBAM), which identifies the target object in its bounding box and thus serves as pseudo ground-truth for weakly supervised semantic and instance segmentation. This approach significantly outperforms recent comparable techniques on both the PASCAL VOC and MS COCO benchmarks in weakly supervised semantic and instance segmentation. In addition, we provide a detailed analysis of our method, offering deeper insight into the behavior of the BBAM.

/pdf/bbam-bounding-box-attribution-map-for-weakly-supervised-d1whi4ywg0.pdf

BBAM: Bounding Box Attribution Map for Weakly Supervised Semantic and Instance Segmentation

MXene nanomaterials stand out among 2D nanomaterials and have been extensively studied by researchers because of their unique layered structure, chemical diversity, and outstanding chemical and physical properties. In recent years, MXene materials have rapidly opened the market for electromagnetic interference shielding technology due to their excellent electromagnetic wave absorption (EMA) capability. However, so far, compared with the development of electromagnetic (EM) shielding technology, there is still much room for development in the construction of MXenes and the application of electromagnetic wave absorption. Herein, the construction strategies, electromagnetic wave conversion mechanism, and applications of MXenes materials are reviewed. Through meticulous material design and an interdisciplinary approach, 2D MXene materials are expected to become one of the smart tunable wave absorbers, and their application scenarios will also become broader.

2D MXene Nanomaterials: Synthesis, Mechanism, and Multifunctional Applications in Microwave Absorption

Morphology control of eco-friendly chitosan-derived carbon aerogels for efficient microwave absorption at thin thickness and thermal stealth

Intelligently detecting and identifying liquids leakage combining triboelectric nanogenerator based self-powered sensor with machine learning

Silver nanowire (Ag NW) has been considered as the promising building block for the fabrication of transparent electromagnetic interference (EMI) shielding films. However, the practical application of Ag NW-based EMI shielding films has been restricted due to the unsatisfactory stability of Ag NW. Herein, we proposed a reduced graphene oxide (rGO) decorated Ag NW film, which realizes a seamless integration of optical transparency, highly efficient EMI shielding, reliable durability and stability. The Ag NW constructs a highly transparent and conductive network, and the rGO provides additional conductive path, showing a superior EMI shielding effectiveness (SE) of 33.62 dB at transmittance of 81.9%. In addition, the top rGO layer enables the hybrid film with reliable durability and chemical stability, which can maintain 96% and 90% EMI SE after 1000 times bending cycles at radius of 2 mm and exposure in air for 80 days. Furthermore, the rGO/Ag NW films also possess fast thermal response and heating stability, making them highly applicable in wearable devices. The synergy of Ag NW and rGO grants the hybrid EMI shielding film multiple desired functions and meanwhile overcomes the shortcomings of Ag NW. This work provides a reference for preparing multifunctional integrated transparent EMI shielding film. 

https://link.springer.com/content/pdf/10.1007/s40820-022-00810-y.pdf

Multifunctional Integrated Transparent Film for Efficient Electromagnetic Protection

/pdf/multifunctional-integrated-transparent-film-for-efficient-2ir7eig0.pdf

Tooth segmentation acts as a crucial and fundamental role in dentistry for doctors to make diagnosis and treatment plans. In this paper, we propose a Two-Stage Attention Segmentation Network (TSASNet) on dental panoramic X-ray images to address the issues suffered in the tooth boundary and tooth root segmentation task which are caused by the low contrast and uneven intensity distribution. We firstly adopt an attention model which is embedded with global and local attention modules to roughly localize the tooth region in the first stage. Without any interactive operator, the attention model so constructed can automatically aggregate pixel-wise contextual information and identify coarse tooth boundaries. To better obtain final boundary information, we use a fully convolutional network as the second stage to further segment the real tooth area from the attention maps obtained from the first stage. The effectiveness of TSASNet is substantiated on the benchmark dataset containing 1,500 dental panoramic X-ray images, our proposed method achieves 96.94% of accuracy, 92.72% of dice and 93.77% of recall, significantly superior to the current state-of-the-art methods.

TSASNet: Tooth segmentation on dental panoramic X-ray images by Two-Stage Attention Segmentation Network

Dynamic texture recognition by aggregating spatial and temporal features via ensemble SVMs

With the popularity of face recognition technology, people have put forward higher requirements for the security of face recognition system. Face anti-spoofing detection attracts extensive attention and many methods been proposed. However, these methods perform poorly in cross scenes. To solve this problem, we propose a face anti-spoofing detection algorithm based on domain adaptation. We apply Maximum Mean Discrepancy (MMD) to multi-layer network distribution adaptation, which improves the generalization ability of the model. To further improve the performance of face anti-spoofing detection, we fuse the low-level features with the high-level features of convolutional neural network for face anti-spoofing detection. Two widely used datasets are used to test the proposed method. The experimental results show that the proposed algorithm outperforms state-of-the-art approaches.

Feng Yang

Papers

Multifunctional Integrated Transparent Film for Efficient Electromagnetic Protection

Multifunctional Integrated Transparent Film for Efficient Electromagnetic Protection

TSASNet: Tooth segmentation on dental panoramic X-ray images by Two-Stage Attention Segmentation Network

Dynamic texture recognition by aggregating spatial and temporal features via ensemble SVMs

Face Anti-Spoofing Based on Multi-layer Domain Adaptation