It is widely believed that increasing the amount of training data enhances the intelligence of deep learning models, which in turn heightens dependence on external datasets. However, these datasets are susceptible to adversarial poisoning, allowing attackers to insert backdoors that trigger misclassifications. Although various defense strategies have been suggested, training-phase defenses (TPDs) appear most promising, as they can significantly lower the attack success rate (ASR) without greatly impacting model performance. Nevertheless, creating TPDs that achieve both high accuracy and low ASR is challenging due to two main issues: 1) Many solutions require additional clean samples that match the distribution of the poisoned dataset, which is not always practical in real-world scenarios; 2) most existing solutions have high computational costs, sometimes requiring five to ten times the expense of standard training, which severely limits their practical use. To tackle these challenges, we introduce Confidence Consistency Detection (CCD), an efficient and lightweight training-phase backdoor detection method. CCD is particularly advantageous in situations where clean data is scarce or unavailable, as it completely eliminates the need for external clean samples. Moreover, CCD significantly reduces computational costs to just 25% to 50% of existing solutions (1.7 times the standard training time), providing a notable improvement over current TPD methods. The core innovation of CCD lies in its ability to utilize the high confidence shown by backdoor samples during the early stages of model training for precise detection. Specifically, we initially train a model on a poisoned dataset for a few epochs, followed by intra-class loss fine-tuning to increase sensitivity to poisoned samples. We then create preliminary sets of poisoned and clean samples by assessing the consistency of confidence variations before and after fine-tuning. These sets guide the model training, enabling the detection of high-confidence poisoned samples. Extensive experiments demonstrate that CCD effectively reduces the attack success rate (ASR) to 1.43%, while having a negligible impact on the model’s clean accuracy. In detecting poisoned samples, CCD achieves a 99% true positive rate (TPR) and a 0.033% false positive rate (FPR), setting a new benchmark in the field.

The automatic detection of defects in industrial photovoltaic film is crucial for ensuring the reliability of photovoltaic modules. Key challenges include limited defect samples, interclass feature similarities, and interference from complex backgrounds. Existing deep learning-based methods require large-scale datasets and focus solely on visual data, which limits their effectiveness in few-shot defect detection (FSDD). To address these challenges, we propose Contextual Ensemble Language-Image multimodal Network (CELIN), which enhances FSDD in photovoltaic films by incorporating textual information through prompt tuning. Unlike traditional language-image models that rely on single fixed text prompts, CELIN employs a position-aware context ensemble strategy to integrate position-specific prompt vectors, enabling the model to capture global contextual information and reduce background interference. In addition, a cross class mask method is introduced to differentiate between similar defect categories by blocking interclass interactions during attention computation, thereby minimizing misclassification. Extensive experiments on our own few-shot photovoltaic film defect dataset and various public benchmarks demonstrates that CELIN significantly outperforms existing methods.

This paper tackles the challenge of Clothes-Changing person Re-Identification (CC-ReID) through the dual lenses of model development and dataset creation. Unlike conventional person ReID, which presumes that individuals maintain consistent clothing, CC-ReID acknowledges the frequent changes in attire encountered in real-world scenarios. This task is particularly difficult due to the substantial variations in visual cues such as colors and textures that accompany changes in clothing. We propose a Multi-modal Assisted feature Learning Framework (MAL-F) designed to learn representations that are invariant to color and texture by utilizing RGB, grayscale, and contour images. MAL-F is a versatile framework that can be seamlessly integrated with existing CC-ReID models, significantly enhancing their accuracy. To further reduce the impact of clothing variations, we introduce a novel CC-ReID backbone named ResTNet, which combines ResNet with a Transformer. ResTNet features a Non-Clothing Salient Region (NC-SR) reinforcement transformer module that employs random clothing block drop and spatial activation mapping. These methods direct the model to emphasize key regions unrelated to clothing, enhancing these features with minimal computational overhead. This strategy substantially enhances CC-ReID performance and supports real-time video processing. Furthermore, recognizing the limitations of current small-scale CC-ReID datasets, we have developed two larger datasets: FBCC and Market1501-CC. Extensive experimental results demonstrate that our proposed methodology surpasses existing state-of-the-art methods for CC-ReID in terms of accuracy, robustness, and efficiency.

Digital dentistry has received more attention in the past decade. However, current deep learning-based methods still encounter difficult challenges. The proposal-based methods are sensitive to the localization results due to the lack of local cues, while the proposal-free methods have poor clustering outputs because of the affinity measured by the low-level characteristics, especially in situations of tightly arranged teeth. In this article, we present a novel proposal-based approach to combine objectness and pointwise knowledge in an attention mechanism for point cloud-based tooth instance segmentation, using local information to improve 3D proposal generation and measuring the importance of local points by calculating the center distance. We evaluate the performance of our approach by constructing a Shining3D tooth instance segmentation dataset. The experimental results verify that our approach gives competitive results when compared with the other available approaches.

Summary This paper presented two schemes of parallel 2D discrete wavelet transform (DWT) on Compute Unified Device Architecture graphics processing units. For the first scheme, the image and filter are transformed to spectral domain by using Fast Fourier Transformation (FFT), multiplied and then transformed back to space domain by using inverse FFT. For the second scheme, the image pixels are convolved directly with filters. Because there is no data relevance, the convolution for data points on different positions could be executed concurrently. To reduce data transfer, the boundary extension and down‐sampling are processed during data loading stage, and transposing is completed implicitly during data storage. A similar skill is adopted when parallelizing inverse 2D DWT. To further speed up the data access, the filter coefficients are stored in the constant memory. We have parallelized the 2D DWT for dozens of wavelet types and achieved a speedup factor of over 380 times compared with that of its CPU version. We applied the parallel 2D DWT in a ring artifact removal procedure; the executing speed was accelerated near 200 times compared with its CPU version. The experimental results showed that the proposed parallel 2D DWT on graphics processing units can significantly improve the performance for a wide variety of wavelet types and is promising for various applications. Copyright © 2015 John Wiley & Sons, Ltd.

Traditional image recognition approaches can achieve high performance only when the images have high resolution and superior quality. A new vehicle logo recognition (VLR) method is proposed to treat low-resolution and poor-quality images captured from urban crossings in intelligent transport system, and the proposed approach is based on statistical random sparse distribution (SRSD) feature and multiscale scanning. The SRSD feature is a novel feature representation strategy that uses the correlation between random sparsely sampled pixel pairs as an image feature and describes the distribution of a grayscale image statistically. Multiscale scanning is a creative classification algorithm that locates and classifies a logo integrally, which alleviates the effect of propagation errors in traditional methods by processing the location and classification separately. Experiments show an overall recognition rate of 97.21% for a set of 3370 vehicle images, which showed that the proposed algorithm outperforms classical VLR methods for low-resolution and inferior quality images and is very suitable for on-site supervision in ITSs.

Extracting spatio-temporally consistent segments from a video sequence is a challenging problem due to the complexity of color, motion and occlusions. Most existing spatio-temporal segmentation approaches have inherent difficulties in handling large displacement with significant occlusions . This paper presents a novel framework for spatio-temporal segmentation. With the estimated depth data beforehand by a multi-view stereo technique, we project the pixels to other frames for collecting the boundary and segmentation statistics in a video, and incorporate them into the segmentation energy for spatio-temporal optimization. In order to effectively solve this problem, we introduce an iterative optimization scheme by first initializing segmentation maps for each frame independently, and then link the correspondences among different frames and iteratively refine them with the collected statistics, so that a set of spatio-temporally consistent volume segments are finally achieved. The effectiveness and usefulness of our automatic framework are demonstrated via its applications for 3D reconstruction, video editing and semantic segmentation on a variety of challenging video examples.

SUMMARY Ring artifacts are a common problem in computed tomography, positron emission tomography, magnetic resonance imaging, and synchrotron radiation images. Before further processing the images such as segmentation and quantification, these artifacts have to be removed or suppressed. Otherwise, they may introduce additional errors for the segmentation and subsequent analysis. This paper proposes an improved ring artifact removal method based on biorthogonal wavelet transform, one‐dimensional fast Fourier transform, and Gaussian damping, which is implemented on general‐purpose computing on graphics processing unit with compute unified device architecture. The experimental results show that the proposed algorithms can be speed up several hundred times compared with the previous algorithms on CPU. The significant performance improvement makes the algorithms much more practical in processing large volume of images in real time. Copyright © 2013 John Wiley & Sons, Ltd.