Postdoctoral Position in 3D Vision & Generative AI for Radio Frequency 3D Perception

Full Time

|

Princeton, NJ

|

Yale University

I am hiring postdocs to join my research team at Yale University. This fully funded position offers an exciting opportunity to develop AI-driven approaches for 3D object reconstruction using millimeter-wave (mmWave) and Radio Frequency (RF) signals. The research focuses on advancing machine learning models—such as generative methods, NeRF, and diffusion models—to reconstruct detailed 3D representations from RF sensing.

Unlike cameras and LiDAR, which are restricted to line-of-sight perception, RF signals can penetrate occlusions, enabling the reconstruction of hidden objects and scenes. By leveraging mmWave and RF-based sensing, our goal is to develop robust 3D perception systems that work in challenging environments where conventional vision techniques fail. In this role, you will design and optimize end-to-end ML pipelines to develop new capabilities in 3D reconstruction and scene understanding.

Key Responsibilities:

  • Model Development: Design, implement, and optimize advanced deep learning models (e.g., NeRF, diffusion models, generative models) to extract high-fidelity 3D representations from RF data.

  • Hybrid Methods: Develop hybrid approaches that integrate classical signal processing techniques with modern ML methods to improve reconstruction quality and robustness.

Required Qualifications:

  • A Ph.D. in Computer Science, Electrical Engineering, Computer Engineering or a related field.

  • Hands-on experience implementing 3D computer vision or computer graphics algorithms and training end-to-end ML models.

  • Strong programming and debugging skills in Python

  • Experience with PyTorch and/or TensorFlow.

  • Expertise in generative AI, 3D computer vision, video processing, or image generation.

Additional Desired Expertise:

  • A proven track record of impactful research, or first-authored publications in top-tier AI, Computer Vision, and Graphics venues (e.g., CVPR, ECCV/ICCV, NeurIPS, ICLR, ICML, ICRA, RSS, IROS, ICCP, 3DV, SIGGRAPH).

  • Experience applying ML techniques to non-conventional imaging data (e.g., mmWave, radar, RF sensing) and integrating signal processing or computational imaging methods is a plus but not required.

Position Details & Application:

  • This is a fully funded postdoctoral position for 1-2 years with a competitive salary and benefits, starting July 2025 or later.

  • Interested candidates should submit a CV to tara.boroushaki@yale.edu

Apply Here!