Radio frequency (RF)-based human activity sensing is an essential component of ubiquitous computing, with WiFi sensing providing a practical and low-cost solution for gesture and activity recognition. However, challenges such as manual data collection, multipath interference, and poor cross-domain generalization hinder real-world deployment. Existing data augmentation approaches often neglect the biomechanical structure underlying RF signals. To address these limitations, we present CM-GR, a cross-modal gesture recognition framework that integrates semantic learning with generative modeling. CM-GR leverages 3D skeletal points extracted from vision data as semantic priors to guide the synthesis of realistic WiFi signals, thereby incorporating biomechanical constraints without requiring extensive manual labeling. In addition, dynamic conditional vectors are constructed from inter-subject skeletal differences, enabling user-specific WiFi data generation without the need for dedicated data collection and annotation for each new user. Extensive experiments on the public MM-Fi dataset and our SelfSet dataset demonstrate that CM-GR substantially improves the cross-subject gesture recognition accuracy, achieving gains of up to 10.26% and 9.5%, respectively. These results confirm the effectiveness of CM-GR in synthesizing personalized WiFi data and highlight its potential for robust and scalable gesture recognition in practical settings.
Text-based visual question answering (TextVQA) requires reasoning over images containing rich textual content, often involving knowledge beyond what is directly observable. Existing methods fuse visual objects and OCR tokens but struggle when questions require external knowledge. Moreover, naively incorporating retrieved knowledge often introduces irrelevant or misleading information, which may hinder reasoning rather than support it. To address these challenges, we propose a TextVQA framework that integrates external prior knowledge to support multimodal reasoning. Given an image and question, a task-aware knowledge retrieval module selects relevant candidates, which are then filtered and verified by a knowledge verification module leveraging large language models. The verified knowledge and question are compressed into compact embeddings via a perceiver-based semantic resampler and jointly processed with visual and OCR features in a multimodal reasoning module. Experiments on the TextVQA and ST-VQA datasets demonstrate that our approach effectively leverages external knowledge to improve performance on knowledge-intensive questions.
Although generative models have achieved impressive visual plausibility in image inpainting, they still face challenges in high-fidelity content restoration, as the generated structures may deviate from real-world appearances. In this paper, to achieve accurate inpainting results, we propose the Focus-and-Disperse Inpainting (FDInpaint), a fine-to-broad context aggregation network that follows a focus-and-disperse paradigm. Specifically, FDInpaint first focuses on accurately identifying and reconstructing subtle details and textures through optimized gradient propagation, and subsequently disperses these refined cues across the entire image to seamlessly integrate restored regions with known content, thereby maintaining global visual consistency. Firstly, considering that image textures are intrinsically characterized by local intensity variations and geometric transitions, gradient information provides essential structural priors for accurately recovering missing content. By capturing complementary edge variations and curvature changes, gradient propagation effectively guides the reconstruction of sharp boundaries, delicate textures, and coherent structures, while reducing ambiguity and artifacts introduced during content generation. According to that, we design the Multi-Granularity Gradient-based Detail Perceptron (MGDP) within a scale-cross augmentation framework to explicitly localize and restore subtle micro-structures that are difficult to recover through semantic features alone. Furthermore, we propose the Patch-based Hybrid Consistency Aggregation (PHCA), a hierarchical consistency propagation module that progressively enlarges the optimization region from small local patches to larger contextual patches and ultimately the whole image. Through this patch-to-image aggregation process, PHCA continuously aligns reconstructed content with known regions by modeling local details and global dependencies, thereby achieving seamless transitions and maintaining global visual consistency. This refinement process enables seamless transitions between inpainted and known areas while maintaining global structural coherence without relying on stochastic sampling. Extensive experiments on Places2, CelebA-HQ, ICDAR, and COCO datasets demonstrate that FDInpaint achieves superior restoration accuracy and visual quality compared with state-of-the-art inpainting methods.
Image fusion aims to combine complementary information from multiple source images to generate more comprehensive scene representations. Existing methods primarily rely on the stacking and design of network architectures to enhance the fusion performance, often ignoring the impact of dataset scene bias on model training. This oversight leads the model to learn spurious correlations between specific scenes and fusion weights under conventional likelihood estimation framework, thereby limiting fusion performance. To solve the above problems, this paper first re-examines the image fusion task from the causality perspective, and disentangles the model from the impact of bias by constructing a tailored causal graph to clarify the causalities among the variables in image fusion task. Then, the Back-door Adjustment based Feature Fusion Module (BAFFM) is proposed to eliminate confounder interference and enable the model to learn the true causal effect. Finally, Extensive experiments on three standard datasets prove that the proposed method significantly surpasses state-of-the-art methods in infrared and visible image fusion.