# Simultaneous Semantic and Collision Learning for 6-DoF Grasp Pose Estimation

Source: [Simultaneous Semantic and Collision Learning for 6-DoF Grasp Pose Estimation](<https://www.sota2.com/research/paper/68fa6ade1e209131cea2aef1>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T23:35:16.571Z

Authors: Yiming Li, Tao Kong, Ruihang Chu, Yifeng Li, Peng Wang, Lei Li

Published: 2021-08-05T07:46:48.000Z

## Abstract

Grasping in cluttered scenes has always been a great challenge for robots, due to the requirement of the ability to well understand the scene and object information. Previous works usually assume that the geometry information of the objects is available, or utilize a step-wise, multi-stage strategy to predict the feasible 6-DoF grasp poses. In this work, we propose to formalize the 6-DoF grasp pose estimation as a simultaneous multi-task learning problem. In a unified framework, we jointly predict the feasible 6-DoF grasp poses, instance semantic segmentation, and collision information. The whole framework is jointly optimized and end-to-end differentiable. Our model is evaluated on large-scale benchmarks as well as the real robot system. On the public dataset, our method outperforms prior state-of-the-art methods by a large margin (+4.08 AP). We also demonstrate the implementation of our model on a real robotic platform and show that the robot can accurately grasp target objects in cluttered scenarios with a high success rate. Project link: https://openbyterobotics.github.io/sscl

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Grasp Detection on GraspNet-1Billion RealSense Novel](<https://www.sota2.com/research/sota/grasp-detection-on-graspnet-1billion-realsense-novel.md>): Grasp Detection; GraspNet-1Billion RealSense Novel; AP: 14.01.

- [Grasp Detection on GraspNet-1Billion RealSense Similar](<https://www.sota2.com/research/sota/grasp-detection-on-graspnet-1billion-realsense-similar.md>): Grasp Detection; GraspNet-1Billion RealSense Similar; AP: 0.2836.

- [Grasp Detection on GraspNet-1Billion RealSense (Seen)](<https://www.sota2.com/research/sota/grasp-detection-on-graspnet-1billion-realsense-seen.md>): Grasp Detection; GraspNet-1Billion RealSense (Seen); AP: 36.55.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2108.02425>)

[PDF](<https://arxiv.org/pdf/2108.02425.pdf>)