Dealing with Ambiguity in Robotic Grasping via Multiple Predictions

Ghazal Ghazaei, Iro Laina, Christian Rupprecht, Federico Tombari, Nassir Navab, Kianoush Nazarpour

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract / Description of output

Humans excel in grasping and manipulating objects because of their life-long experience and knowledge about the 3D shape and weight distribution of objects. However, the lack of such intuition in robots makes robotic grasping an exceptionally challenging task. There are often several equally viable options of grasping an object. However, this ambiguity is not modeled in conventional systems that estimate a single, optimal grasp position. We propose to tackle this problem by simultaneously estimating multiple grasp poses from a single RGB image of the target object. Further, we reformulate the problem of robotic grasping by replacing conventional grasp rectangles with grasp belief maps, which hold more precise location information than a rectangle and account for the uncertainty inherent to the task. We augment a fully convolutional neural network with a multiple hypothesis prediction model that predicts a set of grasp hypotheses in under 60 ms, which is critical for real-time robotic applications. The grasp detection accuracy reaches over 90backslash%for unseen objects, outperforming the current state of the art on this task.
Original languageEnglish
Title of host publicationComputer Vision -- ACCV 2018
EditorsC.V. Jawahar, Hongdong Li, Greg Mori, Konrad Schindler
Place of PublicationCham
PublisherSpringer International Publishing
Number of pages18
ISBN (Electronic)978-3-030-20870-7
ISBN (Print)978-3-030-20869-1
Publication statusPublished - 25 May 2019
Event14th Asian Conference on Computer Vision - Perth, Australia
Duration: 4 Dec 20186 Dec 2018

Publication series

NameLecture Notes in Computer Science
PublisherSpringer, Cham
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Conference14th Asian Conference on Computer Vision
Abbreviated titleACCV 2018
Internet address


Dive into the research topics of 'Dealing with Ambiguity in Robotic Grasping via Multiple Predictions'. Together they form a unique fingerprint.

Cite this