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 Tools annual

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 monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

The demand for accurate food quantification has increased in the recentyears, driven by the needs of applications in dietary monitoring. At the sametime, computer vision approaches have exhibited great potential in automatingtasks within the food domain. Traditionally, the development of machinelearning models for these problems relies on training data sets withpixel-level class annotations. However, this approach introduces challengesarising from data collection and ground truth generation that quickly becomecostly and error-prone since they must be performed in multiple settings andfor thousands of classes. To overcome these challenges, the paper presents aweakly supervised methodology for training food image classification andsemantic segmentation models without relying on pixel-level annotations. Theproposed methodology is based on a multiple instance learning approach incombination with an attention-based mechanism. At test time, the models areused for classification and, concurrently, the attention mechanism generatessemantic heat maps which are used for food class segmentation. In the paper, weconduct experiments on two meta-classes within the FoodSeg103 data set toverify the feasibility of the proposed approach and we explore the functioningproperties of the attention mechanism.

Food Image Classification and Segmentation with Attention-based Multiple Instance Learning