10 November 2021 in IEEE Transactions on Circuits and Systems for Video Technology
Convolutional neural networks (CNNs) are good at extracting contexture features within certain receptive fields, while transformers can model the global long-range dependency features. By absorbing the advantage of transformer and the merit of CNN, Swin Transformer shows strong feature representation ability. Based on it, we propose a cross-modality fusion model, SwinNet , for RGB-D and RGB-T salient object detection. It is driven by Swin Transformer to extract the hierarchical features, boosted by attention mechanism to bridge the gap between two modalities, and guided by edge information to sharp the contour of salient object. To be specific, two-stream Swin Transformer encoder first extracts multi-modality features, and then spatial alignment and channel re-calibration module is presented to optimize intra-level cross-modality features. To clarify the fuzzy boundary, edge-guided decoder achieves inter-level cross-modality fusion under the guidance of edge features. The proposed model outperforms the state-of-the-art models on RGB-D and RGB-T datasets, showing that it provides more insight into the cross-modality complementarity task.
1 October 2019 in Neurocomputing
29 August 2022 in IEEE Transactions on Circuits and Systems for Video Technology
The High-Resolution Transformer (HRFormer) can maintain high-resolution representation and share global receptive fields. It is friendly towards salient object detection (SOD) in which the input and output have the same resolution. However, two critical problems need to be solved for two-modality SOD. One problem is two-modality fusion. The other problem is the HRFormer output’s fusion. To address the first problem, a supplementary modality is injected into the primary modality by using global optimization and an attention mechanism to select and purify the modality at the input level. To solve the second problem, a dual-direction short connection fusion module is used to optimize the output features of HRFormer, thereby enhancing the detailed representation of objects at the output level. The proposed model, named HRTransNet, first introduces an auxiliary stream for feature extraction of supplementary modality. Then, features are injected into the primary modality at the beginning of each multi-resolution branch. Next, HRFormer is applied to achieve forwarding propagation. Finally, all the output features with different resolutions are aggregated by intra-feature and inter-feature interactive transformers. Application of the proposed model results in impressive improvement for driving two-modality SOD tasks, e.g., RGB-D, RGB-T, and light field SOD.https://github.com/liuzywen/HRTransNet.
1 April 2020 in Neurocomputing
1 August 2021 in Swarm and Evolutionary Computation
4 January 2023 in IEEE Transactions on Network Science and Engineering
Multi-objective evolutionary algorithms (MOEAs) have been proved to be competitive for solving the critical node detection problem (CNDP) on complex networks in both non-cascading scenario and cascading scenario. With the continuous expansion of network scale, the search space of the existing MOEAs will increase exponentially. To this end, we propose an interactive co-evolutionary framework (ICoEF) to solve the multi-objective critical node detection on large-scale complex networks, where a set of local populations as well as a global population are interactively evolved to detect the critical nodes with high qualities. Specifically, we first use a community detection algorithm to divide the original network into multiple local communities, where each local community generates a local population for quickly finding local solutions independently. Then, the latent good solutions can be easily obtained by combining the returned local solutions. After that, one global population generated by the original network can refine these latent good solutions by using the proposed two-stage replacement search strategy. Through the above local and global evolutionary interaction, a set of high-quality solutions can be finally obtained. Experiments on synthetic and real-world networks demonstrate the superiority of the proposed framework over several state-of-the-art baselines in both non-cascading and cascading scenarios.