Image forgery detection and localization has become a research topic of increasing interest with the enormous spread of manipulated images on the Web. Most previous methods related to image forgery focus on one particular attribute or neglect the importance of multi-scale information. Such approaches are considered not suitable for detecting and locating multi-scale splicing forgeries. This paper presents a novel approach that exploits residual attention and integrates multi-scale local and global information to improve detection accuracy. In the proposed method, we first aggregate multi-level convolutional feature maps extracted by the encoder to enrich the feature representations and improve the ability of the model to locate multi-scale forged areas. Then, we design a residual attention block (RAB) to purify the features, which enhances the response of task-related regions and suppresses noise information. Furthermore, a global feature mining block (GFMB) is proposed to capture the long-range dependencies between different regions of the image, enabling the model to handle complex tampering scenarios effectively. The multi-scale splicing forgery regions are precisely detected and located by utilizing the proposed method. The extensive experiments are conducted on three benchmark datasets, CASIA, COLUMB, and NIST’16. Specifically, our model achieves the F1 score of 84.3%, 87.9%, and 80.8% on CASIA, COLUMB, and NIST’16 test sets, respectively, outperforming state-of-the-art methods.

The Sensor-based Human Activity Recognition (SHAR) task seeks to recognize signals collected by various sensors embedded in intelligent devices to assist people in their daily lives. Motivated by the success of deep learning, many researchers are studying combining deep learning with SHAR. The key to implementing sensor-based human activity recognition with deep learning lies in facilitating model performance and maintaining efficiency when the model is performed on resource-constrained devices. To address this challenge, we present an effective sensor signal representation method, termed the Temporal-Spatial Dynamic Convolutional (TS-DyConv) network, to recognize human activity. TS-DyConv aims to dynamically learn the convolutional kernels weighted with attention generated along the temporal and spatial kernel spaces. This way, the TS-DyConv can diversify the kernel to enhance the sensor signal’s recognition capabilities without raising complexity and maintaining efficiency. Extensive experiments conducted on three benchmark SHAR datasets, e.g., OPPORTUNITY, PAMAP2, and USC-HAD, demonstrate the superiority of our method over the deep learning baselines and existing SHAR works.

Detection of image splicing forgery become an increasingly difficult task due to the scale variations of the forged areas and the covered traces of manipulation from post-processing techniques. Most existing methods fail to jointly multi-scale local and global information and ignore the correlations between the tampered and real regions in inter-image, which affects the detection performance of multi-scale tampered regions. To tackle these challenges, in this paper, we propose a novel method based on feature aggregation and region-aware learning to detect the manipulated areas with varying scales. In specific, we first integrate multi-level adjacency features using a feature selection mechanism to improve feature representation. Second, a cross-domain correlation aggregation module is devised to perform correlation enhancement of local features from CNN and global representations from Transformer, allowing for a complementary fusion of dual-domain information. Third, a region-aware learning mechanism is designed to improve feature discrimination by comparing the similarities and differences of the features between different regions. Extensive evaluations on benchmark datasets indicate the effectiveness in detecting multi-scale spliced tampered regions.