<abstract xmlns="http://www.w3.org/1999/xhtml"><p>In the paper, the problem of building semantic image segmentation networks in a more efficient way is considered. Building a network capable of successfully segmenting real-world images does not require a real semantic image segmentation task. At this stage, called prototyping, a toy dataset can be used. Such a dataset can be artificial and thus may not need augmentation for training. Besides, its entries are images of much smaller size, which allows training and testing the network a way faster. Objects to be segmented are one or few convex polygons in one image. Thus, a toy dataset generator is created whose complexity is regulated by the number of edges in a polygon, the maximal number of polygons in one image, the set of scale factors, and the set of probabilities determining how many polygons in a current image are generated. The dataset capacity and image size are concurrently adjustable, although they are much less influential.</p></abstract>

In the paper, the problem of building semantic image segmentation networks in a more efficient way is considered. Building a network capable of successfully segmenting real-world images does not require a real semantic image segmentation task. At this stage, called prototyping, a toy dataset can be used. Such a dataset can be artificial and thus may not need augmentation for training. Besides, its entries are images of much smaller size, which allows training and testing the network a way faster. Objects to be segmented are one or few convex polygons in one image. Thus, a toy dataset generator is created whose complexity is regulated by the number of edges in a polygon, the maximal number of polygons in one image, the set of scale factors, and the set of probabilities determining how many polygons in a current image are generated. The dataset capacity and image size are concurrently adjustable, although they are much less influential.

Generator of a Toy Dataset of Multi-Polygon Monochrome Images for Rapidly Testing and Prototyping Semantic Image Segmentation Networks

Electrical, Control and Communication Engineering

This work is licensed under the Creative Commons Attribution 4.0 Public License.

In the paper, the problem of building semantic image segmentation networks in a more efficient way is considered. Building a network capable of successfully...

Generator of a Toy Dataset of Multi-Polygon Monochrome Images for Rapidly Testing and Prototyping Semantic Image Segmentation Networks

Publié en ligne: 20 mars 2020

Pages: 54 - 61

DOI: https://doi.org/10.2478/ecce-2019-0008

Mots clés
Dataset complexity, Multi-polygon object, Semantic image segmentation, Segmentation network architecture, Toy dataset, Two-class segmentation

© 2019 Vadim Romanuke, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 Public License.

Generator of a Toy Dataset of Multi-Polygon Monochrome Images for Rapidly Testing and Prototyping Semantic Image Segmentation Networks

Publié en ligne: 20 mars 2020

Pages: 54 - 61

DOI: https://doi.org/10.2478/ecce-2019-0008

Mots clésDataset complexity, Multi-polygon object, Semantic image segmentation, Segmentation network architecture, Toy dataset, Two-class segmentation

© 2019 Vadim Romanuke, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 Public License.

Mots clés
Dataset complexity, Multi-polygon object, Semantic image segmentation, Segmentation network architecture, Toy dataset, Two-class segmentation