We estimate three mutually orthogonal vanishing directions points in the following steps:
- Detect straight line segments
- Find intersection points (vanishing point candidates)
- Score and rank vanishing point candidates
- Choose the triplet with the highest combined score that also leads to reasonable camera parameters
- Label the line segments based on the estimate
- cv2
$python main.py "filename"
Input files should be stored in "/input"
Output files will be stored in "/output"
Input:
Output:
Vanishing lines are labeled with three directions, the thrid label indicates irrelavent lines.
see the LICENSE.md file for details
This project is generally implemented based on:
- Mallya, A. & Lazebnik, S. (2015). Learning Informative Edge Maps for Indoor Scene Layout Prediction
- Schwing, A. G. & Urtasun, R. (2012). Efficient Exact Inference for 3D Indoor Scene Understanding
- Hedau, V., Hoiem, D. & Forsyth, D. A. (2009). Recovering the spatial layout of cluttered rooms
- Rother, C. (2000). A New Approach for Vanishing Point Detection in Architectural Environments
- Tardif, J.-P. (2009). Non-iterative approach for fast and accurate vanishing point detection
- Denis, P., Elder, J. H. & Estrada, F. J. (2008). Efficient Edge-Based Methods for Estimating Manhattan Frames in Urban Imagery
Contact me for more detailed report on the implementation.