A lean C++ library for working with point cloud data

A python tool for fitting primitives 3D shapes in point clouds using RANSAC algorithm

[CVPR 2024 - Oral] Matching 2D Images in 3D: Metric Relative Pose from Metric Correspondences

Efficient Global Point-cloud registration

The Graph-Cut RANSAC algorithm proposed in paper: Daniel Barath and Jiri Matas; Graph-Cut RANSAC, Conference on Computer Vision and Pattern Recognition, 2018. It is available at http://openaccess.thecvf.com/content_cvpr_2018/papers/Barath_Graph-Cut_RANSAC_CVPR_2018_paper.pdf

Compute Vanishing points using RANSAC and rectify the image

[ICCV-2025] TurboReg: TurboClique for Robust and Efficient Point Cloud Registration

Create panorama image using invariant features from given set of overlapping images.

A demo that implement image registration by matching SIFT descriptors and appling RANSAC and affine transformation.

:gem: "Marker-less Augmented Reality" with OpenCV and OpenGL.

Image alignment and stitching with MATLAB

Problem Set solutions for the "Introduction to Computer Vision (ud810)" MOOC from Udacity

[RAL 2024] RANSAC Back to SOTA: A Two-Stage Consensus Filtering for Real-Time 3D Registration

An Evaluation of Feature Matchers for Fundamental Matrix Estimation (BMVC 2019)

有空就写点，没空就空着。

RANSAC Template Library

MODS (Matching On Demand with view Synthesis) is algorithm for wide-baseline matching.

[CVPR 2023] Two-view Geometry Scoring Without Correspondences

LiDAR Fusion with Vision

Image Mosaicing or Panorama Creation

Implemented a pipeline for 2D image mosaic and stitching. Feature extraction & matching, Adaptive Non-Maximal Suppression (ANMS), geometric blur, RANSAC

image stitching using SIFT, homography matrix, RANSAC and weighted blending. done in openCV.

MODS with external deep descriptors/detectors

Scripts showcasing filtering techniques applied to point cloud data.

A structure from motion implemention in C++ and accelerated using CUDA

PROSAC algorithm in python

Joint Geometric and Object Segmentation for Indoor Scenes

Using Euclidiean Clustering and RANSAC to detect Objects in Lidar captured Point Clouds (PCDs)

Process LIDAR point cloud data for object detection. Implements RANSAC and Euclidean clustering with KD-Tree

A python node to detect planes from depth image by using RANSAC algorithm. Input/Output from/to ROS topics.