• OpenCV200906

OpenCV 2009.08

•  The summer release, OpenCV 2.0 is done, Released Sept. 30, 2009

• Dates:

  • Feature freeze Aug 15

  • Beta Release Sept. 15

  • Released Sept. 30, 2009

The goals in brief

1. New C++ interface (with backward compatibility),
2. Better Python interface with low-level capabilities.
3. More 2D feature detectors and descriptors;
4. Add 3D vision support:
   • Better stereo support. 3D reconstruction, (If we can ...) image stitching.
   • 3D feature detectors and descriptors.
5. Application section:
   • Human detector
   • Camera location from chessboard/3D bounding box
6. Better optical flow (dense, sparse), particle filtering, deformable templates.
7. New functionality and optimizations in MLL, algorithmic and regression tests, better and more universal haartraining.
8. Optimization and threading.
9. Revised documentation.
10. Highgui enhancements.
11. Enhanced regression tests

... And in more details:

OpenCV Summer Release:

• Dates:

  • Feature freeze July 31

  • Beta Released: Sept 14th.

  • Gold Release Sept 29, 2009

• Keys:

  • Done; In progress; Um, maybe; Left as an exercise for the user to implement

PLANS:

• General Structure:

  • (Probably past Beta) Provide way to build OpenCV statically, then the footprint will be variable and usually much smaller than now

• New Features

  • C++ interface

    • cxcore

    • cv

    • HighGUI

    • I/O for sructures, histograms

  • Better python interface

    • In progress thanks to (James Bowman)

    • Move into OpenCV and deprecate SWIG only version -> selected in CMake either or both

    • Use this to expand unit test functions for OpenCV

  • 2D features

    • HOG

      • Chamfer matching with orientation In progress thanks to Marius Muja

    • Keypoints (mostly thanks to LePetit group's great work in this area)

      • LePetit keypoint detector YAPE * (Yet Another Point Extractor). We have found this to be the most stable detector.

      • C-descriptor ... (Vadim Michael -- on site. Difficulties reproducing results)

      • Fern based planar patch detector from LePetit's group D

      • LePetit one way descriptor (Victor) D

    • MSER -- list of contours ... how to extract keypoint from this

    • Fast detector *

    • SURF * D

    • Star *

    • Self-similar detector

  • 3D features

    • Bundle Adjustment (by Phase Space / Sergey)

    • normals in grid (Gary)

    • normal based 3D patch (Gary)

    • SPIN

  • Optical flow

    • Dense (Farnerback)

  • MLL

    • Extreme trees

    • Reduce memory footprint

    • More convenient training via a data class

    • Nister trees (Patrick) Post Beta

    • FLANN or appropriate parts (Marius)

    • Generalized (boosted) cascade (generalizing Haar cascade -- not Haar or LBP features)

• Misc:

  • adaptive skin tracking

• Documentation

  • Move to one Latex source

    • Use to generate HTML, PDF

  • Add search function

  • Add user feedback/documentation bug system

  • Proof read -- Needs complet re-do

  • Put up

    • cxcore

    • cv

    • MLL

    • Aux

    • HighGUI

  • C++ Interface

    • C++ cxcore

    • cv

    • Aux nothing for new functions

    • HighGUI

    • Needs improvement on class descriptions

  • Create Python documenation

• High GUI

  • In memory .jpg, .png

  • move to C++ interface

  • visualizations, statistical data extensions

• Testing functions

  • Regression tests for all functions at least in python->OpenCv to make these easier to write

  • Bought new server for automated tests and builds

    • creating auto test system -> Usable results 3 weeks

• Define user contribution process:

Put off to Future

• 2D features

  • Scale Invariant Feature Transform with Irregular Orientation Histogram Binning, Yan Cui, Nils Hasler, Thorsten Thorm ̈hlen, Hans-Peter Seidel (Feature_Point_Descriptors_SIFT_Improvement_CuiHasThoSei09igSIFT.pdf)

    • Alpha channel, alternatives to contours RLE
    • New alpha transparency (per pixel) added
    • Introduce functions with advance ways of blending
    • Mask channels
• 3D features
  • Plane fit (delayed for integration with Radu's point cloud library -- more comprehensive approach)
  • PFH, FPFH from Radu,
  • 3D Delaunay,
  • 3D convex hull,
  • mesh generation,
  • mesh simplification/subsampling
  • Extend to using 2D XYZA grid as well as oct/kd trees
• 3D capture
  • Probably by using silhouettes and then texture for cavities (?)
    • Possibly with textures or phased light
  • VSLAM
  • Image stitching
• Optical flow
  • Variational optical flow
• MLL
  • Tiechman's
    • Kernel based gentle boost
    • Feature selection engine
  • Automatic parameter tuning
    • Cross-validation, ROC curve generator
• High GUI
  • move to using QT as the HighGUI interface look and feel
• Misc:
  • Focus detector,
• Lighting/Specularity
  • Look into:

    • Using Specularities for Recognition, Margarita Osadchy, David Jacobs, Ravi Ramamoorthi (stored on my disk as specularities.pdf)

    • Retrieving Multiple Light Sources in the Presence of Specular Reflections and Texture, Pascal Lagger and Pascal Fua (Specular_multiple_sources_Fua08a-1.pdf)

    • 3D Pose Refinement from Reflections, Pascal Lagger, Mathieu Salzmann, Vincent Lepetit, Pascal Fua (Specular_lagger_cvpr08.pdf)

    • Detecting Specular Surfaces on Natural Images, Andrey DelPozo, Silvio Savarese

• Histograms
  • Insert value function with interpolation
  • Automatic bin size adjustment for recognition performance using a test set

• OpenCV Ideas 2009

Older Summer 2009 List

Examples

Here's the new proposed STL/Template matrix interface for OpenCV being used for debugging SVD (Note that we will preserve the old interface to OpenCV).

    // matrix A comes in
    SVD svd(A);
    int n = A.rows;

    double a = norm(svd.u*svd.u.t() - Mat::eye(n,n), CV_C);
    double b = norm(svd.vt*svd.vt.t() - Mat::eye(n,n), CV_C);
    int nz = countNonZero(svd.w >= 0);
    double c = norm(svd.u*Mat::diag(svd.w)*svd.vt - Mat::eye(n,n), CV_C);
    CV_Assert( a < 1e-5 && b < 1e-5 && c < 1e-5 && nz == n );

• OpenCV Development

• OpenCV Internal Development