TernausNetV2: Fully Convolutional Network for Instance Segmentation

By Vladimir I. Iglovikov, Selim S. Seferbekov, Alexander V. Buslaev, Alexey A. Shvets. 2018

160 citations recorded by the source corpus.

The most common approaches to instance segmentation are complex and use two-stage networks with object proposals, conditional random-fields, template matching or recurrent neural networks. In this work we present Ternaus- NetV2 - a simple fully convolutional network that allows extracting objects from a high-resolution satellite imagery on an instance level. The network has popular encoderdecoder type of architecture with skip connections but has a few essential modifications that allows using for semantic as well as for instance segmentation tasks. This approach is universal and allows to extend any network that has been successfully applied for semantic segmentation to perform instance segmentation task. In addition, we generalize network encoder that was pre-trained for RGB images to use additional input channels. It makes possible to use transfer learning from visual to a wider spectral range. For DeepGlobe-CVPR 2018 building detection sub-challenge, based on public leaderboard score, our approach shows superior performance in comparison to other methods.

Subfields: Computer Vision and Pattern Recognition, Media Technology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Media Technology

Citations in the corpus

Incremental Learning for Semantic Segmentation of Large-Scale Remote Sensing Data (2018, methodology): “Variants of this network [5]–[7], (e.g., U-net, including VGG-11 [8] encoder and corresponding decoder) have been applied to remote sensing images and have shown a remarkable performance.”
Object Detection and Image Segmentation with Deep Learning on Earth Observation Data: A Review-Part I: Evolution and Recent Trends (2020, background): “The U-Net design was modified, for instance in Tiramisu [68], which combines U-Net structure and DenseNet [46] building blocks, and in U-Net++ [125], where nested dense building blocks perform convolution on the skip connections.”
Angiodysplasia Detection and Localization Using Deep Convolutional Neural Networks (2018, background): “These results can be improved even further by, for example, employing deeper and/or wider encoder architectures [36] and applying more sophisticated image augmentations [37].”
Breast Tumor Cellularity Assessment using Deep Neural Networks (2019, methodology): “In particular, Iglovikov et al. [28, 29] used batch normalization [30] and exponential linear unit (ELU) as the primary activation function and an ImageNet pre-trained VGG-11 network [31] as an encoder.”
Feature Pyramid Network for Multi-class Land Segmentation (2018, background): “Speciﬁc applications of convolutional networks on remote sensing image segmentation have been recently studied [10, 11, 16, 23, 4].”
Camera Model Identification Using Convolutional Neural Networks (2018, background): “Problems in satellite [1], [2], [3], [4], medical [5], [6], [7], [8] or any other imagery are successfully tackled with these techniques, routinely beating human performance in many tasks, including classiﬁcation, segmentation, and detection.”
Automatic Instrument Segmentation in Robot-Assisted Surgery Using Deep Learning (2018, unknown): “These results can be improved even further by, for example, employing deeper and/or wider encoder architectures [24] and applying sophisticated image”
Fully Convolutional Network for Automatic Road Extraction from Satellite Imagery (2018, background): “This type of network architecture proven themselves in binary image segmentation competitions for satellite image analyses [6, 7, 8, 15, 16] and other applications [13, 12].”