Products
Software, apps, datasets, and models from the Baskar Group.
Software, apps, datasets, and models from the group, cross-linked to research themes and papers.
Software
Generate editable 3D plant models from natural language: parametric, analysis-ready geometry for phenotyping and simulation.
Models
A stationary-camera neural radiance field for hyperspectral 3D reconstruction of produce, enabling automated phenotyping without moving cameras.
A latent diffusion model that generates diverse 3D multiphase microstructures in seconds, with controllable volume fractions and tortuosities.
NeRF-based 3D point-cloud reconstruction from a single stationary camera for high-throughput plant phenotyping.
Benchmarks
An expert-curated agricultural reasoning benchmark with 44.6K question-answer pairs, plus small reasoning models that run on consumer GPUs.
Datasets
AutoSiQ: a curated haploid Arabidopsis thaliana inflorescence dataset with a fine-grained silique ontology and a deep learning application for haploid fertil...
Binary 2D Morphologies of Polymer Phase Separation: Dataset and Python Toolbox
162M biodiversity images across 366K species with taxonomic metadata, plus CLIP models and benchmarks. A NeurIPS 2024 Datasets and Benchmarks spotlight.
A large-scale benchmark of fluid flow simulations over complex geometries, so scientific machine learning models train and test on realistic flow physics.
A large-scale multiphase-flow dataset for scientific machine learning: droplet and bubble dynamics.
AI-ready 3D point clouds of field-grown maize with organ-level segmentation, NURBS leaf models, and multiple resolutions.
Live links point to the group's public repositories, project pages, datasets, and demos.