Homography Estimation in Complex Topological Scenes

TL;DR

This paper introduces an automated camera calibration method using a dictionary-based approach with a Spatial Transformer Network and a novel topological loss, improving accuracy in complex scenes without prior camera info.

Contribution

It presents a new automated calibration technique that does not require prior camera knowledge, utilizing a custom STN and a topological loss function for better accuracy.

Findings

Improves IoU metric by up to 12% over state-of-the-art methods

Effective across synthetic datasets and real-world data (World Cup 2014)

Enhances calibration robustness in complex topological scenes

Abstract

Surveillance videos and images are used for a broad set of applications, ranging from traffic analysis to crime detection. Extrinsic camera calibration data is important for most analysis applications. However, security cameras are susceptible to environmental conditions and small camera movements, resulting in a need for an automated re-calibration method that can account for these varying conditions. In this paper, we present an automated camera-calibration process leveraging a dictionary-based approach that does not require prior knowledge on any camera settings. The method consists of a custom implementation of a Spatial Transformer Network (STN) and a novel topological loss function. Experiments reveal that the proposed method improves the IoU metric by up to 12% w.r.t. a state-of-the-art model across five synthetic datasets and the World Cup 2014 dataset.