I’m a PhD candidate of Computer Science at George Washington University. I’m advised by Dr. Robert Pless, with a research focus on understanding images cross long time period and its application in agriculture.

About my work

I am a researcher specializing in deep learning and computer vision. My current focus is on a project that involves analyzing hundreds of sorghum cultivars using a large dataset curated from TERRA.
Utilizing advanced deep metric learning methods, I have constructed an image feature space to help interpret the various phenotypes and genotypes of the sorghum plants. Through visualizing the activation areas of the images, I can better comprehend the relationship between different cultivars and learn about their growth patterns. To better understand this complex data, I have also developed a embedding visualization tool that visualize data in high-dimensional space and enables me to explore different kinds of labels in multimodal datasets. This has helped me identify trends within each cultivar cluster and pinpoint which cultivars do not cluster well at the start of the season.
Prior to this project, I worked on sorghum leaf morphological phenotyping from 3D scanner data. I built a semi-automated annotation tool to label leaves, which was used to train a deep convolutional neural network to segment leaves. I also used the point cloud data to build a leaf voxel adjacency graph to extract phenotypes such as leaf length and curvature.
Through my work in deep learning and computer vision, I believe that we can make significant strides in the agricultural industry by gaining a better understanding of how genotype and environment interact. With my research in sorghum cultivar phenotyping and analysis, I am excited to explore the potential of these technologies to improve agricultural practices. I am passionate about continuing to contribute to this field and am always looking for new research opportunities to further advance our knowledge in this area.

Publications

• SG×P: A Sorghum Genotype×Phenotype Prediction Dataset and Benchmark
  Zeyu Zhang, Robert Pless, Nadia Shakoor, Austin Carnahan, Abby Stylianou NeurIPS 2023 [paper]
• What Does Learning About Time Tell About Outdoor Scenes?
  Zeyu Zhang, Callista Baker, Noor Azam-Naseeruddin, Jingzhou Shen, Robert Pless
  AIPR Workshop 2022
• Comparing Deep Learning Approaches for Understanding Genotype× Phenotype Interactions in Biomass Sorghum.
  Zeyu Zhang, Madison Pope, Nadia Shakoor, Robert Pless, Todd C. Mockler, Abby Stylianou
  Frontiers in Artificial Intelligence 5 2022 [paper]
• Metric Learning for Large Scale Agricultural Phenotyping.
  Zeyu Zhang, Abby Stylianou, Robert Pless
  NAPPN 2021 [paper]
• 2-MAP: Aligned Visualizations for Comparison of High-Dimensional Point Sets”
  Xiaotong Liu, Zeyu Zhang, Hong Xuan, Roxana Leontie, Abby Stylianou, Robert Pless
  WACV 2020 [paper]
• Visualizing Data Driven Phenotypes
  Zeyu Zhang, Hong Xuan, Xiaotong Liu, Abby Stylianou, Robert Pless
  CVPPP 2019 [paper]
• Visualizing how embeddings generalize
  Xiaotong Liu, Hong Xuan, Zeyu Zhang, Abby Stylianou, Robert Pless
  ICML Workshop 2019 [paper]