Reeshad Khan, Ph.D.

Efficient AI (EdgeAI) & Autonomous Systems Perception, Prediction & Planning — BEV • Radar–Vision Fusion • RAW-to-Task Co-Design

GitHub Google Scholar CV (PDF) LinkedIn
Real-Time BEV EdgeAI Deployment Radar–Vision Fusion RAW-to-Task Co-Design 3D Perception, Prediction and Planning
Reeshad Khan headshot

About

I work on efficient perception systems for autonomy with an emphasis on real-time deployment: BEV perception, multisensor fusion, and sensing-aware learning under noisy and resource-constrained settings.

Currently
Ph.D. (CS), University of Arkansas
Dissertation defended Dec 8, 2025

News

acceptances • awards • milestones
2026-01-01
✨
Paper accepted: Adaptive Extensions of Unbiased Risk Estimators for Unsupervised MRI Denoising (CVC 2026).
2025-12-08
🎓
Successfully defended Ph.D. dissertation: “Efficient Deep Neural Networks for Autonomous Perception.”
2025-01-01
🚗
Paper published: TinyBEV (ICCV WDFM 2025).
2025-01-01
🧠
Paper published: From Noise Estimation to Restoration (VISAPP 2025).
2025-01-01
🧾
Paper published: Learning From Oversampling (IEEE Access 2024).
2024-01-01
🏅
Award: Reginald R. “Barney” & Jameson A. Baxter Graduate Fellowship (2024).
2023-01-01
🏅
Award: EECS Graduate Fellowship, University of Arkansas (2023).
2022-01-01
🏅
Award: College of Engineering Graduate Fellowship, University of Arkansas (2022).
2019-01-01
🌏
Scholarship: Chinese Government Belt and Road Scholarship (2019).

Research

Efficient 3D Perception & BEV Autonomy
Resource-aware BEV perception/planning, distillation, and modular pipelines for autonomy.
Robust Learning under Noisy Sensing
Sensing-aware learning, RAW-to-task pipelines, and restoration methods for challenging conditions.

Research Areas

Computer Vision for Autonomous Systems
  • 3D object detection, tracking, motion prediction
  • BEV perception, prediction & planning
  • Safety-aware evaluation and benchmarking
Multisensor Fusion & Robust Perception
  • Radar–vision fusion
  • Robust learning under noisy sensing
  • Uncertainty-aware inference and evaluation
Efficient / Deployable AI (EdgeAI)
  • Resource-constrained models for real-time autonomy
  • Optimization for GPU / embedded deployment
  • Systems-minded pipelines (Linux, Docker, MLOps)
Simulation & Tooling
  • CARLA, SUMO, Unity
  • Experiment design, evaluation, visualization
  • Reproducible training and benchmarking

Certifications

NVIDIA — Accelerated Computing with CUDA Python
Jan 2025
NVIDIA — Generative AI with Diffusion Models
Sep 2025

Reviewer Service

  • Reviewer, ICASSP 2026
  • Reviewer, IEEE Big Data 2022

Publications

click a title to open the PDF

Contact

Email: rk010@uark.edu
LinkedIn: linkedin.com/in/reeshad-khan-a50207106
GitHub: github.com/Reeshad-Khan

Visitors

World visitor map + country counts.

World Map
Country Counter
Flag Counter