# TernausNetV2: Fully Convolutional Network for Instance Segmentation

arXiv: https://arxiv.org/abs/1806.00844 · HTML: https://mlsanity.com/paper/1806.00844/

## Key facts

- Cited by 25 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 Jun 2018.

## Built on by (most cited first, 25 of 25)

- [Continental-Scale Building Detection from High Resolution Satellite Imagery](https://mlsanity.com/paper/2107.12283/) (2021)
- [PolyWorld: Polygonal Building Extraction with Graph Neural Networks in Satellite Images](https://mlsanity.com/paper/2111.15491/) (2021)
- [Fast building segmentation from satellite imagery and few local labels](https://mlsanity.com/paper/2206.05377/) (2022)
- [3D Convolutional Neural Networks for Stalled Brain Capillary Detection](https://mlsanity.com/paper/2104.01687/) (2021)
- [Accurate Polygonal Mapping of Buildings in Satellite Imagery](https://mlsanity.com/paper/2208.00609/) (2022)
- [DeepDamageNet: A two-step deep-learning model for multi-disaster building damage segmentation and classification using satellite imagery](https://mlsanity.com/paper/2405.04800/) (2024)
- [Fully convolutional Siamese neural networks for buildings damage assessment from satellite images](https://mlsanity.com/paper/2111.00508/) (2021)
- [Lebanon Solar Rooftop Potential Assessment using Buildings Segmentation from Aerial Images](https://mlsanity.com/paper/2111.11397/) (2021)
- [Cross-dimensional transfer learning in medical image segmentation with deep learning](https://mlsanity.com/paper/2307.15872/) (2023)
- [GridTracer: Automatic Mapping of Power Grids using Deep Learning and Overhead Imagery](https://mlsanity.com/paper/2101.06390/) (2021)
- [Sci-Net: Scale Invariant Model for Buildings Segmentation from Aerial Imagery](https://mlsanity.com/paper/2111.06812/) (2021)
- [H-EMD: A Hierarchical Earth Mover's Distance Method for Instance Segmentation](https://mlsanity.com/paper/2206.01309/) (2022)
- [MARS: Mask Attention Refinement with Sequential Quadtree Nodes for Car Damage Instance Segmentation](https://mlsanity.com/paper/2305.04743/) (2023)
- [Transparency Distortion Robustness for SOTA Image Segmentation Tasks](https://mlsanity.com/paper/2405.12864/) (2024)
- [Investigating transformers in the decomposition of polygonal shapes as point collections](https://mlsanity.com/paper/2108.07533/) (2021)
- [Effect of the output activation function on the probabilities and errors in medical image segmentation](https://mlsanity.com/paper/2109.00903/) (2021)
- [Uncertainty, Edge, and Reverse-Attention Guided Generative Adversarial Network for Automatic Building Detection in Remotely Sensed Images](https://mlsanity.com/paper/2112.05335/) (2021)
- [Meta-simulation for the Automated Design of Synthetic Overhead Imagery](https://mlsanity.com/paper/2209.08685/) (2022)
- [Transformers For Recognition In Overhead Imagery: A Reality Check](https://mlsanity.com/paper/2210.12599/) (2022)
- [Influence of Water Droplet Contamination for Transparency Segmentation](https://mlsanity.com/paper/2405.12861/) (2024)
- [On-Sensor Binarized Fully Convolutional Neural Network with A Pixel Processor Array](https://mlsanity.com/paper/2202.00836/) (2022)
- [Fast and Precise Binary Instance Segmentation of 2D Objects for Automotive Applications](https://mlsanity.com/paper/2208.11527/) (2022)
- [Automatic satellite building construction monitoring](https://mlsanity.com/paper/2209.15084/) (2022)
- [Segment Anything in Defect Detection](https://mlsanity.com/paper/2311.10245/) (2023)
- [Automated National Urban Map Extraction](https://mlsanity.com/paper/2404.06202/) (2024)

## Abstract (arXiv)

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 TernausNetV2 - a simple fully convolutional network that allows extracting objects from a high-resolution satellite imagery on an instance level. The network has popular encoder-decoder 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. The source code corresponding pre-trained weights are publicly available at https://github.com/ternaus/TernausNetV2