Search Results for author: Florent Lafarge

Found 8 papers, 0 papers with code

SCR: Smooth Contour Regression with Geometric Priors

no code implementations8 Feb 2022 Gaetan Bahl, Lionel Daniel, Florent Lafarge

While object detection methods traditionally make use of pixel-level masks or bounding boxes, alternative representations such as polygons or active contours have recently emerged.

Instance Segmentation object-detection +2

Finding Good Configurations of Planar Primitives in Unorganized Point Clouds

no code implementations CVPR 2022 Mulin Yu, Florent Lafarge

We present an algorithm for detecting planar primitives from unorganized 3D point clouds.

Road Extraction from Overhead Images with Graph Neural Networks

no code implementations9 Dec 2021 Gaetan Bahl, Mehdi Bahri, Florent Lafarge

By contrast, we propose a method that directly infers the final road graph in a single pass.

graph construction

Planar Shape Detection at Structural Scales

no code implementations CVPR 2018 Hao Fang, Florent Lafarge, Mathieu Desbrun

Interpreting 3D data such as point clouds or surface meshes depends heavily on the scale of observation.

KIPPI: KInetic Polygonal Partitioning of Images

no code implementations CVPR 2018 Jean-Philippe Bauchet, Florent Lafarge

Recent works showed that floating polygons can be an interesting alternative to traditional superpixels, especially for analyzing scenes with strong geometric signatures, as man-made environments.


Image Partitioning Into Convex Polygons

no code implementations CVPR 2015 Liuyun Duan, Florent Lafarge

The over-segmentation of images into atomic regions has become a standard and powerful tool in Vision.

Recovering Line-Networks in Images by Junction-Point Processes

no code implementations CVPR 2013 Dengfeng Chai, Wolfgang Forstner, Florent Lafarge

Our experiments on a variety of problems illustrate the potential of our approach in terms of accuracy, flexibility and efficiency.

Computer Vision Point Processes

Cannot find the paper you are looking for? You can Submit a new open access paper.