A Survey of 6D Object Detection Based on 3D Models for Industrial Applications

Gorschlüter, Felix; Rojtberg, Pavel; Pöllabauer, Thomas

doi:10.3390/jimaging8030053

Cited by 10 publications

(3 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A simple strategy is to copy objects from real images and then paste them onto random background images to create new images [ 13 , 14 ]. For industrial applications, available 3D models can be used to train object detection models [ 15 ]. Domain randomization is an approach where training images are randomized to such an extent that the trained model is supposed to see real images as just another variation of the synthetic training data [ 5 , 16 , 17 ].…”

Section: Related Workmentioning

confidence: 99%

Combining Synthetic Images and Deep Active Learning: Data-Efficient Training of an Industrial Object Detection Model

Eversberg,

Lambrecht

2024

J. Imaging

View full text Add to dashboard Cite

Generating synthetic data is a promising solution to the challenge of limited training data for industrial deep learning applications. However, training on synthetic data and testing on real-world data creates a sim-to-real domain gap. Research has shown that the combination of synthetic and real images leads to better results than those that are generated using only one source of data. In this work, the generation of synthetic training images via physics-based rendering is combined with deep active learning for an industrial object detection task to iteratively improve model performance over time. Our experimental results show that synthetic images improve model performance, especially at the beginning of the model’s life cycle with limited training data. Furthermore, our implemented hybrid query strategy selects diverse and informative new training images in each active learning cycle, which outperforms random sampling. In conclusion, this work presents a workflow to train and iteratively improve object detection models with a small number of real-world images, leading to data-efficient and cost-effective computer vision models.

show abstract

Section: Related Workmentioning

confidence: 99%

Combining Synthetic Images and Deep Active Learning: Data-Efficient Training of an Industrial Object Detection Model

Eversberg,

Lambrecht

2024

J. Imaging

View full text Add to dashboard Cite

show abstract

“…Pose estimation is a widely applied task in computer vision since the 1980s by Fischler et al [10] and other disciplines that require, for example, the automatic tracking of objects or single-or multiple-view projection based estimation, i.e., see Groschlüteret al [11], He et al [12] and Sahin et al [19] for recent detailed reviews of state-of-the-art techniques. The goal is always to find the position and orientation of a predefined 3D object, which is observed typically by cameras.…”

Section: Introductionmentioning

confidence: 99%

A Stochastic-Geometrical Framework for Object Pose Estimation Based on Mixture Models Avoiding the Correspondence Problem

Hoegele

2024

J Math Imaging Vis

View full text Add to dashboard Cite

Pose estimation of rigid objects is a practical challenge in optical metrology and computer vision. This paper presents a novel stochastic-geometrical modeling framework for object pose estimation based on observing multiple feature points. This framework utilizes mixture models for feature point densities in object space and for interpreting real measurements. Advantages are the avoidance to resolve individual feature correspondences and to incorporate correct stochastic dependencies in multi-view applications. First, the general modeling framework is presented, second, a general algorithm for pose estimation is derived, and third, two example models (camera and lateration setup) are presented. Numerical experiments show the effectiveness of this modeling and general algorithm by presenting four simulation scenarios for three observation systems, including the dependence on measurement resolution, object deformations and measurement noise. Probabilistic modeling utilizing mixture models shows the potential for accurate and robust pose estimations while avoiding the correspondence problem.

show abstract

“…Three highly valuable survey papers were accepted in this Special Issue. Gorschlüter et al [ 5 ] present a survey on 6DoF object detection and pose estimation, which has become a key topic of AR and robotic applications in recent years. They provide an industrial application perspective and focus their review on the methods that exclusively use synthetic data from 3D models of the object for training.…”

mentioning

confidence: 99%