Applied Machine Learning Engineer with a PhD in Computer Vision. I work on ML systems across data generation, training pipelines, model adaptation, evaluation, and optimization for deployment.

My public work includes research projects on synthetic data, domain adaptation, semantic segmentation, and simulation for computer vision. My production ML work has mostly been in private or company-owned codebases.

You can find the full list on Google Scholar.

• SPIN: An Open Simulator of Realistic Spacecraft Navigation Imagery [arXiv]
  Unity-based simulator for generating synthetic spacecraft navigation imagery.

• SynthmanticLiDAR: A Synthetic Dataset for Semantic Segmentation on LiDAR Imaging [arXiv]
  Synthetic LiDAR dataset for semantic segmentation and sim-to-real experimentation.

• Leveraging Contrastive Learning for Semantic Segmentation with Consistent Labels Across Varying Appearances [arXiv]
  Domain adaptation work using synthetic data and contrastive learning for semantic segmentation.

• Unsupervised Class Generation to Expand Semantic Segmentation Datasets [arXiv]
  Dataset expansion approach using generative models and segmentation tools.

• Exploiting semantic segmentation to boost reinforcement learning in video game environments [MTAP]
  Using semantic segmentation as an intermediate representation for reinforcement learning.

More details about my publications and projects are available at jmontalvo.dev.