Learning deep domain-agnostic features from synthetic renders for industrial visual inspection

Abubakr, Abdelrahman G.; Jovančević, Igor; Mokhtari, Nour Islam; Abdallah, Hamdi Ben; Orteu, Jean-José

doi:10.1117/1.jei.31.5.051604

Cited by 3 publications

(1 citation statement)

References 62 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the study from Abdelrahman et al., 20 the authors generated synthetic 2D images of mechanical parts with random real background images from industrial settings. To produce these images, they experimented with various rendering algorithms.…”

Section: Previous Workmentioning

confidence: 99%

Domain shared features for visual inspection of complex mechanical assemblies based on synthetically generated data

Došljak,

Jovančević,

Orteu

2024

J. Electron. Imag.

Self Cite

View full text Add to dashboard Cite

Even though neural network methodologies have been established for a long time, only recently have they achieved exceptional efficacy in practical deployments, predominantly due to improvements in hardware computational capacity and the large amounts of available data for learning. Nonetheless, substantial challenges remain in utilizing deep learning in many domains, mainly because of the lack of large amounts of labeled data that are versatile enough for deep learning models to learn useful information. For instance, in mechanical assembly inspection, annotating data for each type of mechanical part to train a deep learning model can be very labor-intensive. Additionally, it is required to annotate data after each modification of mechanical part specification. Also, the system for inspection is typically not available until the first few samples are built to collect data. This paper proposes a solution for these challenges in case of the visual mechanical assembly inspection by processing point cloud data acquired via a three-dimensional (3D) scanner. To reduce the necessity for manually labeling large amounts of data, we employed synthetically generated data for both training and validation purposes, reserving the real sensor data exclusively for the testing phase. Our approach reduces the need for large amounts of labeled data by using synthetically generated point clouds from computer-aided design models for neural network training. Domain gap is a significant challenge for the usage of synthetically generated data. To reduce the domain gap, we used different preprocessing techniques, as well as a neural network architecture that focuses more on shared features that will not significantly change between synthetically generated data and real data from the 3D sensor.

show abstract

Section: Previous Workmentioning

confidence: 99%

Domain shared features for visual inspection of complex mechanical assemblies based on synthetically generated data

Došljak,

Jovančević,

Orteu

2024

J. Electron. Imag.

Self Cite

View full text Add to dashboard Cite

show abstract

Towards cognition-augmented human-centric assembly: A visual computation perspective

Pang,

Zheng,

Fan

et al. 2025

Robotics and Computer-Integrated Manufacturing

View full text Add to dashboard Cite

A Novel Deep Learning-Based Visual Search Engine in Digital Marketing for Tourism E-Commerce Platforms

Wu,

Liu

2024

Journal of Organizational and End User Computing

View full text Add to dashboard Cite

Visual search technology, because of its convenience and high efficiency, is widely used by major tourism e-commerce platforms in product search functions. This study introduces an innovative visual search engine model, namely CLIP-ItP, aiming to thoroughly explore the application potential of visual search in tourism e-commerce. The model is an extension of the CLIP (contrastive language-image pre-training) framework and is developed through three pivotal stages. Firstly, by training an image feature extractor and a linear model, the visual search engine labels images, establishing an experimental visual search engine. Secondly, CLIP-ItP jointly trains multiple text and image encoders, facilitating the integration of multimodal data, including product image labels, categories, names, and attributes. Finally, leveraging user-uploaded images and jointly selected product attributes, CLIP-ItP provides personalized top-k product recommendations.

show abstract

Learning deep domain-agnostic features from synthetic renders for industrial visual inspection

Cited by 3 publications

References 62 publications

Domain shared features for visual inspection of complex mechanical assemblies based on synthetically generated data

Domain shared features for visual inspection of complex mechanical assemblies based on synthetically generated data

Towards cognition-augmented human-centric assembly: A visual computation perspective

A Novel Deep Learning-Based Visual Search Engine in Digital Marketing for Tourism E-Commerce Platforms

Contact Info

Product

Resources

About