End-to-End Learning of Semantic Grasping

TL;DR

This paper introduces an end-to-end framework for semantic robotic grasping that integrates object recognition and grasp planning, leveraging self-supervised data and semi-supervised learning to improve performance.

Contribution

It presents a novel two-stream neural network architecture inspired by visual reasoning theories, combining object detection and grasping in a unified model trained with minimal supervision.

Findings

End-to-end training outperforms non-end-to-end systems.

Semi-supervised learning enhances grasping accuracy.

Autonomous data collection reduces need for manual labeling.

Abstract

We consider the task of semantic robotic grasping, in which a robot picks up an object of a user-specified class using only monocular images. Inspired by the two-stream hypothesis of visual reasoning, we present a semantic grasping framework that learns object detection, classification, and grasp planning in an end-to-end fashion. A "ventral stream" recognizes object class while a "dorsal stream" simultaneously interprets the geometric relationships necessary to execute successful grasps. We leverage the autonomous data collection capabilities of robots to obtain a large self-supervised dataset for training the dorsal stream, and use semi-supervised label propagation to train the ventral stream with only a modest amount of human supervision. We experimentally show that our approach improves upon grasping systems whose components are not learned end-to-end, including a baseline method that uses bounding box detection. Furthermore, we show that jointly training our model with auxiliary data consisting of non-semantic grasping data, as well as semantically labeled images without grasp actions, has the potential to substantially improve semantic grasping performance.