Lukas Roth

This paper uses neural processes with normalizing flows to model wheat height trajectories, evaluating their ability to capture distributional dynamics influenced by genotype and environment.

OneGrow generates temporal plant images and organ masks, addressing the lack of paired data by jointly modeling wheat images and masks over time.

This paper introduces a deep learning model that integrates phenomic, genomic, and environmental data to predict wheat yield across diverse environments, showing improved performance over traditional genomic models.

A 1,096-image pixel-labelled wheat organ dataset plus 52,078 unlabelled field images from 11 institutions enables Segformer leaf and spike segmentation near 90% mIoU, while stems lag.

The paper uses deep learning to track lesion growth in wheat infected by Zymoseptoria tritici, finding lesion growth is a significant component of quantitative resistance in most cultivars.

The paper introduces a large-scale wheat phenotyping dataset with time series imaging and trait data across six years, enabling studies on genotype-environment interactions and crop modeling.

This study compares multispectral and RGB sensors on UAVs for monitoring wheat canopy senescence, finding that frequent measurements are more effective than high spectral resolution.