Media Summary: Jonathan Tremblay, Stephen Tyree, Terry Mosier, Stan Birchfield. Long version for IROS 2020 paper Jonathan Tremblay, Stephen Tyree, Terry Mosier, Stan Birchfield. For the first time, an algorithm trained only on synthetic data is able to beat a state-of-the-art network trained on real images for ...
Indirect Object To Robot Pose - Detailed Analysis & Overview
Jonathan Tremblay, Stephen Tyree, Terry Mosier, Stan Birchfield. Long version for IROS 2020 paper Jonathan Tremblay, Stephen Tyree, Terry Mosier, Stan Birchfield. For the first time, an algorithm trained only on synthetic data is able to beat a state-of-the-art network trained on real images for ... MIT 6.881 Final Project presentation Fall 2020 Lucy Lee, Yorai Shaoul. Neural network trained only on synthetic data for skeletal Vision-based state and pose estimation for robotic bin picking of cables
For more info, including open source code: Presentation video for the paper "Precise Multi-Modal In-Hand
