Semantic Scholar Open Access 2018 666 sitasi

TernausNet: U-Net with VGG11 Encoder Pre-Trained on ImageNet for Image Segmentation

V. Iglovikov Alexey A. Shvets

Abstrak

Pixel-wise image segmentation is demanding task in computer vision. Classical U-Net architectures composed of encoders and decoders are very popular for segmentation of medical images, satellite images etc. Typically, neural network initialized with weights from a network pre-trained on a large data set like ImageNet shows better performance than those trained from scratch on a small dataset. In some practical applications, particularly in medicine and traffic safety, the accuracy of the models is of utmost importance. In this paper, we demonstrate how the U-Net type architecture can be improved by the use of the pre-trained encoder. Our code and corresponding pre-trained weights are publicly available at this https URL. We compare three weight initialization schemes: LeCun uniform, the encoder with weights from VGG11 and full network trained on the Carvana dataset. This network architecture was a part of the winning solution (1st out of 735) in the Kaggle: Carvana Image Masking Challenge.

Topik & Kata Kunci

Penulis (2)

V

V. Iglovikov

A

Alexey A. Shvets

Format Sitasi

Iglovikov, V., Shvets, A.A. (2018). TernausNet: U-Net with VGG11 Encoder Pre-Trained on ImageNet for Image Segmentation. https://doi.org/10.1007/978-3-030-64340-9_15

Akses Cepat

PDF tidak tersedia langsung

Cek di sumber asli →
Lihat di Sumber doi.org/10.1007/978-3-030-64340-9_15
Informasi Jurnal
Tahun Terbit
2018
Bahasa
en
Total Sitasi
666×
Sumber Database
Semantic Scholar
DOI
10.1007/978-3-030-64340-9_15
Akses
Open Access ✓