Current cross-modal food retrieval approaches focus mainly on the global visual appearance of food without explicitly considering multi-grained information. Additionally, direct calculation of the global similarity of image-recipe pairs is not particularly effective in terms of latent alignment, which suffers from mismatch during the mutual image-recipe retrieval process. This paper proposes a threefold encoder interaction (TEI) cross-modal food retrieval framework to maintain the multi-granularity of food images and the multi-levels of textual recipes to address the aforementioned challenges. The TEI framework comprises an image encoder, a recipe encoder, and a multi-grained interaction encoder. We simultaneously propose a multi-grained relation-aware attention (MRA) embedded in the multi-grained interaction encoder to capture multi-grained food visual features. The multi-grained interaction similarity scores are calculated to better establish the multi-grained correlation between recipe and image entities based on the extracted hierarchical textual and multi-grained visual features. Finally, a hierarchical multi-grained semantic alignment loss is designed to supervise the whole process of cross-modal training using the multi-grained interaction similarity scores. Extensive qualitative and quantitative experiments on the Recipe1M dataset have demonstrated that the proposed TEI framework achieves multi-grained semantic alignment between image and text modalities and is superior to other state-of-the-art methods in cross-modal food retrieval tasks.
21 August 2025 in Expert Systems with Applications
Pavement crack segmentation is crucial for enhancing traffic safety, improving maintenance efficiency, extending road lifespan, and supporting smart city development. Utilising computer vision technology to automate crack detection can significantly reduce time and labour costs, improving both accuracy and efficiency. However, pavement crack images exhibit complex visual features, irregular distributions, and diverse shapes and textures, posing challenges for accurate segmentation. To address these issues, a pavement crack segmentation network (PCSNet) based on a dynamic dual-branch encoder and deformable spatial focusing is proposed. The dual-branch encoder employs pre-trained and self-trained branches to extract general and specific crack features, respectively. Dynamic feature fusion optimises the contribution of each branch, enhancing model generalisation. The deformable spatial focusing module refines crack morphological features, improving the model’s ability to identify and localise cracks of varying shapes. Extensive experiments on the DeepCrack dataset show that PCSNet achieves precision, recall, F1 score, and Mean Intersection over Union of 85.34%, 86.16%, 85.75% and 75.23%, respectively, outperforming all comparative methods, thereby validating its superiority.
22 January 2026 in Expert Systems with Applications