Machine Vision open Source processing library Summary
From Cvchina to get to the machine vision open source processing library Summary, turn to, very force, still constantly updated ...
General Purpose Libraries/general Library
No need to say much.
Recognition and Vision Library. Thread safety. A powerful IO mechanism. Contains AAM.
Cool one image processing package. The entire library has only one header file. Contains a PDE-based optical flow algorithm.
Images, video io/image, videos IO
- Freeimage
- DevIL
- ImageMagick
- Ffmpeg
- Videoinput
- Portvideo
AR related/augmented Reality
AR Library based on marker
Enhanced version of Artoolkit. A better attitude estimation algorithm is implemented.
Real-time tracking, SLAM, AR library. No need for marker, templates, built-in sensors, etc.
AR library based on feature point detection and recognition.
Local invariant features/local invariant Feature
Currently the best sift open source implementation. The Kd-tree,kd-forest,bow implementation is also included.
Feature point recognition based on naive Bayesian bundle. High speed, but high memory consumption.
Sift implementation based on OPENCV.
Target Detection/object Detection
Another adaboost implementation. Training speed is fast.
- Pedestrian detection by Jianxin.wu
Fast pedestrian detection based on centrist and linear SVM.
(approximate) Nearest neighbor/ann
Currently the most complete (approximate) nearest open Source library. It not only realizes a series of search algorithms, but also includes a mechanism to automatically select the quickest algorithm.
Another approximate nearest neighbor library.
SLAM & SFM
Monoslam Library. Developed by Androw Davison.
Image Segmentation/segmentation
Use simple Linear iterative clustering to produce a specified number, approximately evenly distributed super Pixel.
Target Tracking/tracking
Target tracking algorithm based on online Random forest.
Kanade-lucas-tracker
Online Boosting Trackers
Line Detection/line Detection
A straight line detection algorithm based on unicom domain connection.
A gradient-based, local line segment detection operator.
Fingerprint/finger Print
Multimedia file hash algorithm based on perception. (Extract, compare image, video, audio fingerprint)
Visual significance of/visual salience
- Global contrast Based Salient region Detection
Ming-ming Cheng's visual significance algorithm.
FFT/dwt
The fastest, best open source FFT.
A lightweight FFT implementation. The license is the highlight.
Audio Processing/audio Processing
Audio processing, audio synthesis.
Audio file IO.
Audio resampling.
Wavelet transform
Fast wavelet transform (FWT)
Brief:binary Robust Independent Elementary Feature a good local feature descriptor with fast Corner + BRIEF to achieve feature point matching demo:http:// cvlab.epfl.ch/software/brief/
Http://code.google.com/p/javacv
Java packaged OpenCV, FFmpeg, libdc1394, PGR flycapture, Openkinect, Videoinput, and artoolkitplus libraries. Can be put on Android to use ~
Libhik,hik SVM, calculates HIK SVM with centrist Lib. Http://c2inet.sce.ntu.edu.sg/Jianxin/projects/libHIK/libHIK.htm
A set of visual significance detection code Links: http://cg.cs.tsinghua.edu.cn/people/~cmm/saliency/
Http://www.cnblogs.com/CVArt/archive/2011/10/19/2217609.html
Machine Vision open Source processing library Summary