3D reconstruction of seeds; today's reading.

dataset seeds three-dimensional-reconstruction today-i-read

  1. Vsevolod Cherepashkin, Erenus Yildiz, Andreas Fischbach, Leif Kobbelt, and Hanno Scharr, “Deep Learning Based 3d Reconstruction for Phenotyping of Wheat Seeds: a Dataset, Challenge, and Baseline Method,” in Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshops, Oct. 2023, pp. 561–571.

    tl;dr: Three-dimensional reconstruction of wheat seeds for phenotyping. The most relevant trait is seed volume, because it is indicative of seed mass, which correlates to nutrients available to a seedling plant. The dataset consists of image data from robotic system phenoSeeder; different proportions of data are used in different scenarios. Test data is held back and uses three views; dataset train / val sets and challenge at the website link. Baseline methods are VGG11 and ResNet-152, no code published for the baselines.

© Amy Tabb 2018 - 2023. All rights reserved. The contents of this site reflect my personal perspectives and not those of any other entity.