English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Image-based dietary assessment serves as an efficient and accurate solutionfor recording and analyzing nutrition intake using eating occasion images asinput. Deep learning-based techniques are commonly used to perform imageanalysis such as food classification, segmentation, and portion sizeestimation, which rely on large amounts of food images with annotations fortraining. However, such data dependency poses significant barriers toreal-world applications, because acquiring a substantial, diverse, and balancedset of food images can be challenging. One potential solution is to usesynthetic food images for data augmentation. Although existing work hasexplored the use of generative adversarial networks (GAN) based structures forgeneration, the quality of synthetic food images still remains subpar. Inaddition, while diffusion-based generative models have shown promising resultsfor general image generation tasks, the generation of food images can bechallenging due to the substantial intra-class variance. In this paper, weinvestigate the generation of synthetic food images based on the conditionaldiffusion model and propose an effective clustering-based training framework,named ClusDiff, for generating high-quality and representative food images. Theproposed method is evaluated on the Food-101 dataset and shows improvedperformance when compared with existing image generation works. We alsodemonstrate that the synthetic food images generated by ClusDiff can helpaddress the severe class imbalance issue in long-tailed food classificationusing the VFN-LT dataset.

Diffusion Model with Clustering-based Conditioning for Food Image Generation