Started new role as a Postdoctoral Scientist at NEC Laboratories America, Inc. Media Analytics Team
Ali K. AlShami
Postdoctoral Scientist at NEC Laboratories America, Inc.
Ph.D. in Computer Science with over six years of research experience in Artificial Intelligence (AI), Machine Learning (ML), Computer Vision (CV), and Natural Language Processing (NLP). In addition, five years of industry experience in systems engineering. My mission is to bridge recent advances in AI, CV, ML to develop advanced perception and prediction technologies for autonomous systems, with a focus on safety-critical environments and reliable decisionmaking. In particular my work focuses on design robust foundation models, new methodologies and benchmarks that address novelty problem in different scenarios, leveraging Vision Large Language Models (VLLMs) to enhance contextual understanding and improve automatic data engines to detect objects in complex environments for autonomous system applications.
Research Interests
Foundation Models & Multimodal Reasoning
VLMs, LLMs, and generative models for autonomous driving; vision-language perception and scene understanding; embodied AI for decision-making in AVs.
Open-World & Robust Autonomy
Open-set recognition, OOD detection, and novel hazard avoidance; domain adaptation, transfer learning, and continual learning for robust autonomy.
Prediction, Planning & Interaction
Motion forecasting, trajectory prediction, and behavior modeling; pedestrian intention and human-agent interaction; planning under uncertainty.
Multimodal Perception & Sensor Fusion
Fusing camera, LiDAR, radar, and maps for scene understanding; spatio-temporal representation learning for dynamic environments.
Generative Models & Simulation
Generative models for simulation, data augmentation, and scenario synthesis; synthetic data and sim-to-real transfer.
Systems, Deployment & Evaluation
Efficient training, model compression, and edge deployment; real-time inference; benchmarks and safety-centric evaluation.
News
Our paper “2COOOL: An Evaluation Benchmark for Generating Incident Reports on Out-of-Distribution Hazards in Autonomous Driving” has been accepted to WACV 2026.
Our paper “GATEPose: A Graph Attention Transformer Enhanced with Pose and Orientation Angles for Pedestrian Crossing Intention Prediction” has been accepted to WACV 2026.
Our workshop proposal “AUTOPILOT” has been accepted and will take place at CVPR 2026, Denver, Colorado.
I successfully defended my Ph.D.