# Joint Background Reconstruction and Foreground Segmentation via A Two-stage Convolutional Neural Network

arXiv: https://arxiv.org/abs/1707.07584 · HTML: https://mlsanity.com/paper/1707.07584/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 24 Jul 2017.

## Built on by (most cited first, 4 of 4)

- [An Empirical Review of Deep Learning Frameworks for Change Detection: Model Design, Experimental Frameworks, Challenges and Research Needs](https://mlsanity.com/paper/2105.01342/) (2021)
- [ZBS: Zero-shot Background Subtraction via Instance-level Background Modeling and Foreground Selection](https://mlsanity.com/paper/2303.14679/) (2023)
- [Flow-based Video Segmentation for Human Head and Shoulders](https://mlsanity.com/paper/2104.09752/) (2021)
- [Spatio-temporal Data Augmentation for Visual Surveillance](https://mlsanity.com/paper/2101.09895/) (2021)

## Abstract (arXiv)

Foreground segmentation in video sequences is a classic topic in computer vision. Due to the lack of semantic and prior knowledge, it is difficult for existing methods to deal with sophisticated scenes well. Therefore, in this paper, we propose an end-to-end two-stage deep convolutional neural network (CNN) framework for foreground segmentation in video sequences. In the first stage, a convolutional encoder-decoder sub-network is employed to reconstruct the background images and encode rich prior knowledge of background scenes. In the second stage, the reconstructed background and current frame are input into a multi-channel fully-convolutional sub-network (MCFCN) for accurate foreground segmentation. In the two-stage CNN, the reconstruction loss and segmentation loss are jointly optimized. The background images and foreground objects are output simultaneously in an end-to-end way. Moreover, by incorporating the prior semantic knowledge of foreground and background in the pre-training process, our method could restrain the background noise and keep the integrity of foreground objects at the same time. Experiments on CDNet 2014 show that our method outperforms the state-of-the-art by 4.9%.