Max Robotics — Global Service Robot Directory + US Certification

About

James Hays is a leading researcher in computer vision and robotics, with a primary focus on autonomous driving perception, 3D scene understanding, and robotic manipulation. His most impactful contribution is the creation of **Argoverse**, a landmark dataset for autonomous vehicle research that has garnered over 1,500 citations. Argoverse provides rich, multi-modal sensor data (including 360-degree cameras and LiDAR) from real-world urban environments, along with high-definition maps, enabling breakthroughs in 3D tracking and motion forecasting. Beyond autonomous driving, Hays has made significant strides in robotic manipulation. His work on **ContactPose** and **ContactGrasp** advances functional, multi-finger grasp synthesis, while his **Visual Pressure Estimation and Control** system allows soft robotic grippers to precisely manipulate objects using only visual feedback. Hays also addresses practical deployment challenges through techniques like **multi-teacher progressive distillation** for lightweight object detectors, and **crossmodal transfer learning** to leverage HD maps for improved 3D detection. His research consistently bridges the gap between high-quality academic datasets and real-world robotic systems, making him a pivotal figure in enabling machines to perceive and interact with the physical world.