SOFI: Multi-Scale Deformable Transformer for Camera Calibration with Enhanced Line Queries

TL;DR

SOFI introduces a multi-scale deformable transformer with enhanced line queries for improved camera calibration, leveraging cross-scale interactions to achieve superior accuracy on multiple datasets while maintaining efficiency.

Contribution

The paper presents SOFI, a novel transformer model that enhances line queries with geometric features and incorporates multi-scale deformable attention for camera calibration.

Findings

Outperforms existing methods on multiple datasets

Maintains competitive inference speed

Effectively utilizes cross-scale interactions

Abstract

Camera calibration consists of estimating camera parameters such as the zenith vanishing point and horizon line. Estimating the camera parameters allows other tasks like 3D rendering, artificial reality effects, and object insertion in an image. Transformer-based models have provided promising results; however, they lack cross-scale interaction. In this work, we introduce \textit{multi-Scale defOrmable transFormer for camera calibratIon with enhanced line queries}, SOFI. SOFI improves the line queries used in CTRL-C and MSCC by using both line content and line geometric features. Moreover, SOFI's line queries allow transformer models to adopt the multi-scale deformable attention mechanism to promote cross-scale interaction between the feature maps produced by the backbone. SOFI outperforms existing methods on the \textit {Google Street View}, \textit {Horizon Line in the Wild}, and \textit {Holicity} datasets while keeping a competitive inference speed.