Eric Cosatto

Papers

Total Citations

446

H-Index

About

Eric Cosatto is a leading researcher in machine learning and computer vision, with a particular focus on end-to-end learning for autonomous systems. His most influential contribution is the seminal 2005 paper "Off-Road Obstacle Avoidance through End-to-End Learning," which has garnered over 440 citations. In this groundbreaking work, Cosatto and his team demonstrated that a neural network could be trained directly from raw camera images to steering commands, bypassing traditional hand-crafted perception pipelines. This approach, trained in supervised mode using human driving data collected across diverse off-road terrains, proved remarkably effective for real-time obstacle avoidance. The work laid essential groundwork for modern end-to-end autonomous driving systems, influencing both academic research and industrial applications. Beyond this landmark paper, Cosatto has made significant contributions to medical image analysis and visual recognition, consistently pushing the boundaries of what deep learning can achieve when applied to complex, real-world perception tasks. His research exemplifies the power of learning-based approaches to solve problems that traditional computer vision methods struggle with.