Mirror-image peptides composed of D-amino acids offer exceptional storage density, stability, and longevity, making them a promising medium for biological data storage. However, accurate de-novo sequencing remains challenging due to limited tandem mass spectrometry data and algorithmic constraints. To address this, we propose DBond, a deep neural network that predicts peptide bond cleavage by integrating sequence, precursor ion, and environmental features, enabling the design of mirror-image peptides that are easier to sequence. We further introduce MiPD513, a dataset of 513 mirror-image peptides, and a peptide bond cleavage labeling algorithm (PBCLA) that generates approximately 12.5 million labeled samples. Two prediction strategies, multi-label and single-label classification, are proposed, with the single-label approach found to perform better and providing a basis for sequence optimization. Code is available at https://github.com/LoserLus/DBond.

Carotid plaque classification plays a critical role in the identification of vulnerable plaques, so it is crucial for early risk estimation of cardiovascular and cerebrovascular events. Carotid ultrasound examination with ultrasound images and reports produced by professional doctors is the most common way to assess atherosclerotic plaques in clinical practice. However, existing deep learning methods for carotid ultrasound image analysis ignore the information in the ultrasound report. In this paper, we propose a multi-task learning (MTL) method named NDDR-LCS based on convolutional neural network (CNN) that leverages auxiliary information from ultrasound reports to assist the carotid plaque classification task. NDDR-LCS utilizes dense blocks as feature descriptors and organically combines three novel MTL mechanisms that are Neural Discriminative Dimensionality Reduction (NDDR), Learning Mixtures, and Cross-Stitch, to learn dependencies between ultrasound images and ultrasound reports. Based on carotid ultrasound images and their corresponding diagnostic reports, we conduct sufficient experiments to prove that NDDR-LCS outperforms state-of-the-art CNN methods for carotid plaque classification.

Diversity of training data significantly affects tracking robustness of model under unconstrained environments. However, existing labeled datasets for facial landmark tracking tend to be large but not diverse, and manually annotating the massive clips of new diverse videos is extremely expensive. To address these problems, we propose a Spatial-Temporal Knowledge Integration (STKI) approach. Unlike most existing methods which rely heavily on labeled data, STKI exploits supervisions from unlabeled data. Specifically, STKI integrates spatial-temporal knowledge from massive unlabeled videos, which has several orders of magnitude more than existing labeled video data on the diversity, for robust tracking. Our framework includes a self-supervised tracker and an image-based detector for tracking initialization. To avoid the distortion of facial shape, the tracker leverages adversarial learning to introduce facial structure prior and temporal knowledge into cycle-consistency tracking. Meanwhile, we design a graph-based knowledge distillation method, which distills the knowledge from tracking and detection results, to improve the generalization of the detector. The fine-tuned detector can provide tracker on unconstrained videos with high-quality tracking initialization. Extensive experimental results show that the proposed method achieves state-of-the-art performance on comprehensive evaluation datasets.

Detecting objects at vastly different scales is a fundamental challenge in computer vision. To solve this, some approaches (e.g. TridentNet) investigate the effect of receptive fields, whereas other approaches (e.g. SNIP, SNIPER) are based on the image pyramid strategy. In this study, a novel single‐shot based detector, called scale specified single‐shot multibox detector (4SD) is proposed. It aims to predict objects of a specific scale range separately by using feature maps of different sizes. First, a parallel multi‐branch architecture with feature maps of different sizes is generated by scale specific inference module. Then, the authors propose a scale specific training scheme to specialise each branch by sampling object instances of proper scales for training. Results are shown on both PASCAL VOC and COCO detection. The proposed method can achieve a mean average precision of 83.1% on PASCAL VOC 2007, and 36.9% on MS‐COCO at a speed of 28 frames per second, which is superior to most single‐stage detectors.

Action recognition has been an active research area for many years. Extracting discriminative spatial and temporal features of different actions plays a key role in accomplishing this task. Current popular methods of action recognition are mainly based on two‐stream Convolutional Networks (ConvNets) or 3D ConvNets. However, the computational cost of two‐stream ConvNets is high for the requirement of optical flow while 3D ConvNets takes too much memory because they have a large amount of parameters. To alleviate such problems, the authors propose a Dual Attention ConvNet (DANet) based on dual attention mechanism which consists of spatial attention and temporal attention. The former concentrates on main motion objects in a video frame by using ConvNet structure and the latter captures related information of multiple video frames by adopting self‐attention. Their network is entirely based on 2D ConvNet and takes in only RGB frames. Experimental results on UCF‐101 and HMDB‐51 benchmarks demonstrate that DANet gets comparable results among leading methods, which proves the effectiveness of the dual attention mechanism.

Data augmentation is always a routinely trick in neural network training to improve generalization of a model. However, traditional transformation based methods are domain-specific, the transformations are required to be carefully designed. Recently, Generative Adversarial Networks (GAN) has been proposed to generate new samples which match the real data distribution. But directly using GAN generated samples in data augmentation faces the problems of label absence and uncertain data quality. In this paper, we propose an efficient and robust data augmentation method using GAN generated samples. This method proposes a modified GAN to generate more diverse samples and label them with a soft distribution labeling method. With an improved stochastic gradient descent, all the data are used to train the final classifier. The experiments are conducted on the widely used datasets: MNIST, SVHN and CIFAR-10. Our method empirically obtains promising results, even with few original data.

In this paper, using hierarchical DCT, the notation maximal locally linear similarity (MLL similarity for short) is defined and discussed. The intention to the discussion is that MLL similarity can be used to construct a scheme of the vector quantization coding for medical images and images in other application areas