Point Pair Feature Based Object Detection for Random Bin Picking

Abbeloos, Wim; Goedemé, Toon

doi:10.1109/crv.2016.59

Cited by 16 publications

(6 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The object pose estimation problem in bin picking scenarios has been investigated using local invariant features (e.g., point pair features [1,2]) and template-matching [3]. However, these approaches do not show acceptable performance in bin picking from a cluttered pile of textureless small objects.…”

Section: Related Workmentioning

confidence: 99%

“…To tackle this problem, an RGB-D or depth camera is usually installed on top of the bin. There are existing solutions to bin picking of large objects, mostly using local invariant features [1,2] or template-matching algorithms [3], which rely on the computationally expensive evaluation of many pose hypotheses. Moreover, local features do not perform well for texture-less objects, and thus, template-matching often fails in heavily cluttered scenes with severely occluded objects.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Object detection and Autoencoder-based 6D pose estimation for highly cluttered Bin Picking

Höfer¹,

Shamsafar²,

Benbarka³

et al. 2021

Preprint

View full text Add to dashboard Cite

Bin picking is a core problem in industrial environments and robotics, with its main module as 6D pose estimation. However, industrial depth sensors have a lack of accuracy when it comes to small objects. Therefore, we propose a framework for pose estimation in highly cluttered scenes with small objects, which mainly relies on RGB data and makes use of depth information only for pose refinement. In this work, we compare synthetic data generation approaches for object detection and pose estimation and introduce a pose filtering algorithm that determines the most accurate estimated poses. We will make our real dataset for object detection available with the paper.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Object detection and Autoencoder-based 6D pose estimation for highly cluttered Bin Picking

Höfer¹,

Shamsafar²,

Benbarka³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Some research suggests that primates and humans have separate neural pathways for object recognition and grasping [17], and the object detection and pose estimation has often been treated as an isolated problem separated from grasp selection in the bin-picking literature. Geometry-based methods have been well explored for pose estimation in bin-picking, such as Abbeloos et al [24], that uses the popular point pair feature approach, first presented by Drost et al [27]. Buchhilz et al [26] suggests a two-stage approach where the full object pose is estimated after grasping based on inertial features.…”

Section: Related Workmentioning

confidence: 99%

“…This is a problem that often occurs in industrial settings where objects come out of a production line packaged in bulk, without isolating individual objects, and where the objects are transported to a second production line that subsequently must isolate and process these objects individually. Due to the importance and relevance of the problem, bin picking has been well studied [19], [24]- [26] in the literature. Challenges in bin picking arise when seeking to develop a bin picking algorithm that can be automatically customized for specific objects, and when these objects are very reflective.…”

Section: Introductionmentioning

confidence: 99%

Bin Picking of Reflective Steel Parts Using a Dual-Resolution Convolutional Neural Network Trained in a Simulated Environment

Dyrstad

Bakken

Grøtli

et al. 2018

2018 IEEE International Conference on Robotics and Biomimetics (ROBIO)

View full text Add to dashboard Cite

We consider the case of robotic bin picking of reflective steel parts, using a structured light 3D camera as a depth imaging device. In this paper, we present a new method for bin picking, based on a dual-resolution convolutional neural network trained entirely in a simulated environment. The dualresolution network consists of a high resolution focus network to compute the grasp and a low resolution context network to avoid local collisions.The reflectivity of the steel parts result in depth images that have a lot of missing data. To take this into account, training of the neural net is done by domain randomization on a large set of synthetic depth images that simulate the missing data problems of the real depth images. We demonstrate both in simulation and in a real-world test that our method can perform bin picking of reflective steel parts.

show abstract

“…• Point pair feature similarity: Point pair features [13][8] [3] describe the relative position and orientation of points on the surface of an object ( Figure 2). For two points m 1 and m 2 with normals n 1 and n 2 ,…”

Section: Matching Triplets Of Keypointsmentioning

confidence: 99%

3D Object Discovery and Modeling Using Single RGB-D Images Containing Multiple Object Instances

Abbeloos

Ataer-Cansızoğlu

Caccamo

et al. 2017

2017 International Conference on 3D Vision (3DV)

Self Cite

View full text Add to dashboard Cite

Unsupervised object modeling is important in robotics, especially for handling a large set of objects. We present a method for unsupervised 3D object discovery, reconstruction, and localization that exploits multiple instances of an identical object contained in a single RGB-D image. The proposed method does not rely on segmentation, scene knowledge, or user input, and thus is easily scalable. Our method aims to find recurrent patterns in a single RGB-D image by utilizing appearance and geometry of the salient regions. We extract keypoints and match them in pairs based on their descriptors. We then generate triplets of the keypoints matching with each other using several geometric criteria to minimize false matches. The relative poses of the matched triplets are computed and clustered to discover sets of triplet pairs with similar relative poses. Triplets belonging to the same set are likely to belong to the same object and are used to construct an initial object model. Detection of remaining instances with the initial object model using RANSAC allows to further expand and refine the model. The automatically generated object models are both compact and descriptive. We show quantitative and qualitative results on RGB-D images with various objects including some from the Amazon Picking Challenge. We also demonstrate the use of our method in an object picking scenario with a robotic arm.

show abstract

Point Pair Feature Based Object Detection for Random Bin Picking

Cited by 16 publications

References 33 publications

Object detection and Autoencoder-based 6D pose estimation for highly cluttered Bin Picking

Object detection and Autoencoder-based 6D pose estimation for highly cluttered Bin Picking

Bin Picking of Reflective Steel Parts Using a Dual-Resolution Convolutional Neural Network Trained in a Simulated Environment

3D Object Discovery and Modeling Using Single RGB-D Images Containing Multiple Object Instances

Contact Info

Product

Resources

About