# Temporally Consistent Horizon Lines

## TL;DR

This paper introduces a novel CNN architecture with an improved residual convolutional LSTM and an adaptive loss function for more accurate and temporally consistent horizon line estimation in videos, supported by an extended KITTI dataset.

## Contribution

It presents a new CNN model with residual convolutional LSTM for video horizon line estimation and extends the KITTI dataset with precise horizon labels.

## Key findings

- The proposed method outperforms existing approaches in accuracy.
- Temporal consistency improves horizon line estimation stability.
- The extended KITTI dataset provides valuable data for future research.

## Abstract

The horizon line is an important geometric feature for many image processing and scene understanding tasks in computer vision. For instance, in navigation of autonomous vehicles or driver assistance, it can be used to improve 3D reconstruction as well as for semantic interpretation of dynamic environments. While both algorithms and datasets exist for single images, the problem of horizon line estimation from video sequences has not gained attention. In this paper, we show how convolutional neural networks are able to utilise the temporal consistency imposed by video sequences in order to increase the accuracy and reduce the variance of horizon line estimates. A novel CNN architecture with an improved residual convolutional LSTM is presented for temporally consistent horizon line estimation. We propose an adaptive loss function that ensures stable training as well as accurate results. Furthermore, we introduce an extension of the KITTI dataset which contains precise horizon line labels for 43699 images across 72 video sequences. A comprehensive evaluation shows that the proposed approach consistently achieves superior performance compared with existing methods.