Research bookmarks

Selected libraries and tools

A curated set of bookmarks across image processing, computer vision, machine learning, segmentation, object detection, and 3D reconstruction.

General image processing

OpenCV[BSD]: Image and matrix manipulation, transforms, feature extraction.
CImg[GPL/LGPL]: Open-source C++ toolkit for image processing.
Point Cloud Library (PCL)[BSD]: Filtering, feature estimation, surface reconstruction, registration, and segmentation for point clouds.
SimpleCV[MIT]: Python interface to several open-source computer vision libraries.
BoofCV[Apache]: Java library for real-time computer vision and robotics.

Image acquisition, decoding and encoding

FFmpeg[LGPL/GPL]: Record, convert, and stream audio and video.
ImageMagick[Apache]: Read and write a wide range of image formats.
DevIL[LGPL]: Loading and saving common image formats.
FreeImage[GPL/FPL]: PNG, BMP, JPEG, TIFF loading.

Segmentation

OpenCV — pyramid segmentation[BSD]: Pyramid-based image segmentation.

Machine learning

Torch[BSD]: Gradient machines, SVMs, ensemble models, K-nearest neighbors, GMMs, HMMs, Bayes classifier.

Object detection

OpenCV — Viola–Jones[BSD]: Viola–Jones face detection with Haar features.
INRIA Object Detection Toolkit[Custom]: Histograms of Oriented Gradients for object detection.

Statistical shape modeling

Statismo[BSD]: Framework for shape and image modeling using point distribution models.

3D reconstruction and registration

CMVS / PMVS: Patch-based multi-view stereo software.
libpointmatcher[BSD]: Modular ICP library for 2D and 3D points.
Open3D[MIT]: 3D data processing — visualization, ICP, mesh and point-cloud algorithms.

This list is a representative selection. For a wider list of tools used across Real3d projects, see the research resources page.