Removing people from complex backgrounds in real time
Person removal from complex backgrounds over time. Removing people from complex backgrounds in real-time using TensorFlow.js in the web browser using JavaScript. This code attempts to learn over time the makeup of the background of a video such that I can attempt to remove any humans from the scene. This is all happening in real-time, in the browser, using TensorFlow.js. This is an experiment. It may not be perfect in all situations. Go ahead and try it right now in your own web browser. Feel...
Shape recognition for video games using fast algorithms (heuristics).
This project is a really small software that can be used as a demonstration of my own shape recognition algorithms.
There are a process during the shape drawing and a post-processing when it is done. Finally the shape recognition is launched.
All the algorithms are done for real time gaming. So they are heuristics and do not give the optimal solution. Yet the given solution is generally a good one.
In this demonstration, few shapes can be recognized:
* circles
* triangles
* lines...
VisAmp is a visually controlled mp3 player. It was initially developed during the "Softwarepraktikum" at the Chair for Image Processing and Pattern Recognition of the University of Freiburg, Germany in 2001.
The Mimas Toolkit is a C++ real-time computer vision library. Algorithms include edge/corner-detection, object recognition/tracking, LSI-filters, segmentation, array-operators, convolution etc. OO wrappers for LAPACK, libxine, V4L, FFTW are provided.