Our first contribution is to introduce CRISP, a category-agnostic object pose and shape estimation pipeline. The pipeline implements an encoder-decoder model for shape estimation. It uses ...

Abstract: The 6D object pose estimation task, which determines the 3D position and orientation of an object in the camera coordinate system, is critical for applications such as robotic manipulation, ...

This project provides a complete, open-source pipeline for estimating the real-world position and orientation of FRC game pieces from camera images. It uses a data-driven image-matching approach, ...

The proposed framework will enable robots to accurately and more efficiently handle complex objects, while also advancing augmented reality technologies to support more lifelike hand-object ...

Estimating the pose of hand-held objects is a critical and challenging problem in robotics and computer vision. While leveraging multi-modal RGB and depth data is a promising solution, existing ...