Year: 2024
Type: preprint
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.
Source: arXiv (Cornell University)
Authors: Janampa, Sebastian, Pattichis, Marios
Institution: University of New Mexico
Language: English
Topic: Advanced Vision and Imaging
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Funder: National Science Foundation
Awards: Developing and Testing Bilingual Curricula that Infuse Authentic Computer Programming Experiences into Middle School Mathematics for Latinx Youth, Computer-Assisted Video Analysis Methods for Understanding Underrepresented Student Participation and Learning in Collaborative Learning Environments, Broadening Participation of Latina/o Students in Engineering Using an Integrated Mathematics, Engineering and Computing Curriculum in Authentic, Out-of-School Environments
OpenAlex ID: W4403786357