# Scoring Graspability based on Grasp Regression for Better Grasp Prediction

arXiv: https://arxiv.org/abs/2002.00872 · HTML: https://mlsanity.com/paper/2002.00872/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 Feb 2020.

## Built on by (most cited first, 4 of 4)

- [End-to-end Trainable Deep Neural Network for Robotic Grasp Detection and Semantic Segmentation from RGB](https://mlsanity.com/paper/2107.05287/) (2021)
- [NBMOD: Find It and Grasp It in Noisy Background](https://mlsanity.com/paper/2306.10265/) (2023)
- [Double-Dot Network for Antipodal Grasp Detection](https://mlsanity.com/paper/2108.01527/) (2021)
- [Lightweight Convolutional Neural Network with Gaussian-based Grasping Representation for Robotic Grasping Detection](https://mlsanity.com/paper/2101.10226/) (2021)

## Abstract (arXiv)

Grasping objects is one of the most important abilities that a robot needs to master in order to interact with its environment. Current state-of-the-art methods rely on deep neural networks trained to jointly predict a graspability score together with a regression of an offset with respect to grasp reference parameters. However, these two predictions are performed independently, which can lead to a decrease in the actual graspability score when applying the predicted offset. Therefore, in this paper, we extend a state-of-the-art neural network with a scorer that evaluates the graspability of a given position, and introduce a novel loss function which correlates regression of grasp parameters with graspability score. We show that this novel architecture improves performance from 82.13% for a state-of-the-art grasp detection network to 85.74% on Jacquard dataset. When the learned model is transferred onto a real robot, the proposed method correlating graspability and grasp regression achieves a 92.4% rate compared to 88.1% for the baseline trained without the correlation.