Blog
About

1
views
0
recommends
+1 Recommend
1 collections
    0
    shares
      • Record: found
      • Abstract: found
      • Article: found
      Is Open Access

      Synthetic dataset generation for object-to-model deep learning in industrial applications

      , , , , ,

      PeerJ Computer Science

      PeerJ

      Read this article at

      ScienceOpenPublisher
      Bookmark
          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

          Abstract

          The availability of large image data sets has been a crucial factor in the success of deep learning-based classification and detection methods. Yet, while data sets for everyday objects are widely available, data for specific industrial use-cases (e.g., identifying packaged products in a warehouse) remains scarce. In such cases, the data sets have to be created from scratch, placing a crucial bottleneck on the deployment of deep learning techniques in industrial applications. We present work carried out in collaboration with a leading UK online supermarket, with the aim of creating a computer vision system capable of detecting and identifying unique supermarket products in a warehouse setting. To this end, we demonstrate a framework for using data synthesis to create an end-to-end deep learning pipeline, beginning with real-world objects and culminating in a trained model. Our method is based on the generation of a synthetic dataset from 3D models obtained by applying photogrammetry techniques to real-world objects. Using 100K synthetic images for 10 classes, an InceptionV3 convolutional neural network was trained, which achieved accuracy of 96% on a separately acquired test set of real supermarket product images. The image generation process supports automatic pixel annotation. This eliminates the prohibitively expensive manual annotation typically required for detection tasks. Based on this readily available data, a one-stage RetinaNet detector was trained on the synthetic, annotated images to produce a detector that can accurately localize and classify the specimen products in real-time.

          Related collections

          Most cited references 5

          • Record: found
          • Abstract: not found
          • Book Chapter: not found

          Microsoft COCO: Common Objects in Context

            Bookmark
            • Record: found
            • Abstract: not found
            • Conference Proceedings: not found

            Learning Deep Object Detectors from 3D Models

              Bookmark
              • Record: found
              • Abstract: not found
              • Conference Proceedings: not found

              Render for CNN: Viewpoint Estimation in Images Using CNNs Trained with Rendered 3D Model Views

                Bookmark

                Author and article information

                Journal
                PeerJ Computer Science
                PeerJ
                2376-5992
                2019
                October 14 2019
                : 5
                : e222
                Article
                10.7717/peerj-cs.222
                © 2019

                https://creativecommons.org/licenses/by/4.0/

                Product
                Self URI (article page): https://peerj.com/articles/cs-222

                Comments

                Comment on this article