background preloader

Coding Robin

Coding Robin
Open this page, allow it to access your webcam and see your face getting recognized by your browser using JavaScript and OpenCV, an "open source computer vision library". That's pretty cool! But recognizing faces in images is not something terribly new and exciting. Wouldn't it be great if we could tell OpenCV to recognize something of our choice, something that is not a face? That is totally possible! Here's the good news: we can generate our own cascade classifier for Haar features. But now for the best of news: keep on reading! The following instructions are heavily based on Naotoshi Seo's immensely helpful notes on OpenCV haartraining and make use of his scripts and resources he released under the MIT licencse. Let's get started The first thing you need to do is clone the repository on GitHub I made for this post. git clone You'll also need OpenCV on your system. If you're on OS X and use homebrew it's as easy as this: find . Related:  Graphics and ArtInteractives

memory consumption while training > 50GB answered Jul 5 '13 Prasanna It may not be an answer as such but its too long for a comment. So here it goes. CreateSamples You can actually use this utility in four ways. Creating Distorted Training samples - Perspective Transform , Intensity Variation, etc. Traincascade It's pretty much covered in the OpenCV Documentation. Regarding Size and to know why it blows up go check the haarfeatures.cpp in $OpenCV_DIR/apps/traincascade/ and check the "void CvHaarEvaluator::generateFeatures()" and try calculating how many "sums" you are creating per image and add the size of integral images to that also. And well, the following is obvious. Phew, that was long. Things to note - Use small sample size for creating samples. The pipeline I used : Get distorted samples from original images and generate vec file for each image using createsamples utility Merge the vec files into one ( and check them if you want. Hope this helps. Regards, Prasanna S

Image Segmentation pff's homepage Below is a C++ implementation of the image segmentation algorithm described in the paper: Efficient Graph-Based Image Segmentation Pedro F. The source code is available as a tgz file segment.tgz, or zip segment.zip (updated on 3/21/07). Example segmentation results: Segmentation parameters: sigma = 0.5, K = 500, min = 50. Segmentation parameters: sigma = 0.5, K = 1000, min = 100. Bracelet - The Future is Now (A) These general website terms and conditions of use (hereinafter referred to as the “General Terms and Conditions”) set forth the terms and conditions applicable to the website (hereinafter referred to as the “Site”). (B) The Site is the exclusive property of CN2P (hereinafter referred to as the “Company”). (C) The purpose of the Site is to present the project developed by the Company (the “Project”) and to enable the collection of donations through the Site in order to finance the completion of the Company’s Project (the “Donations”). 2.1 Definitions For the purposes hereof, capitalized terms shall have the meaning given thereto below, unless the context requires otherwise: “Company” means CN2P, a société par actions simplifiée incorporated under the laws of France, whose registered office is La Grande Arche Paroi Nord 92044 Paris La Défense, registered under number RCS Nanterre B 808 557 573; 2.2 Interpretation (ii) plural terms include the singular and vice versa;

Amazon.com: Counting Tray with Spatula: Health & Personal Care Tutorial: OpenCV haartraining (Rapid Object Detection With A Cascade of Boosted Classifiers Based on Haar-like Features) - Naotoshi Seo Tutorial: OpenCV haartraining (Rapid Object Detection With A Cascade of Boosted Classifiers Based on Haar-like Features) Objective The OpenCV library provides us a greatly interesting demonstration for a face detection. However, I could not follow how OpenCV developers performed the haartraining for their face detection system exactly because they did not provide us several information such as what images and parameters they used for training. My working environment is Visual Studio + cygwin on Windows XP, or on Linux. FYI: I recommend you to work haartrainig with something different concurrently because you have to wait so many days during training (it would possibly take one week). A picture from the OpenCV website History 10/16/2008 - Additional experimental results.08/28/2008 - Revised entirely.06/05/2007 - opencv-1.0.003/12/2006 - First Edition (opencv-0.9.7) Tag: SciSoftware ComputerVision FaceDetection OpenCV Data Prepartion . and Computer Vision Test Images Positive (Face) Images . 1.

Quickposes: pose generator for figure & gesture drawing practice 'Ninja Run' May Be the Craziest VR Locomotion Technique Yet This is ‘Ninja Run VR’, and its yet another take on the ever present problem of user locomotion in VR experiences. But, as crazy as the concept looks, there may be some method to its madness. The question of how best to allow users of VR apps, games and experiences to move around a virtual space is still a hot topic among players and developers. Now, a developer has come up with a new take on the problem, one that on the surface may well look, well, a little “batshit crazy”. “In many popular anime and cartoons, characters depicted to be moving very fast are often illustrated with their hands and arms trailing their torso in the direction they are moving,” Hall tells us, “Even though is is unrealistic, it does induce the notion that the person has a super power to move very fast.” The developer has begun to flesh out his concepts too, this has resulted in a demo experience which adopts the technique.

Starting at $99 - People Counters & Door Counters RTC-P3 People Counter: The RTC-P3 people counter gives you an enormous amount of flexibility. The RTC-P3 is a modular unit, not an all-in-one unit. This means that the RTC-P3 people counter and door sensors are housed in separate enclosures. The RTC-P3 people counter can be placed anywhere; however, it is usually placed behind the sales counter or in the back office, near the PC or POS to which the RTC-P3 is connected. The RTC-P3 can accept up to three door sensors. The RTC-S2X Standard Door Sensor and the RTC-S2XL Long-Range Door Sensor, which are sold separately and which connect to the RTC-P3 people counter via Category 5 Networking Cable. The RTC-P3 can be connected to a PC or POS, but it is not required. The standard door sensor that we use with the RTC-P3, the RTC-S2X, can sense traffic on entrances as wide as approximately eight feet in directional mode and approximately 13 feet in nondirectional mode.

Computer Vision Software » Blog Archive » FAQ: OpenCV Haartraining Hi All, before posting your question, please look at this FAQ carefully! Also you can read OpenCV haartraining article. If you are sure, there is no answer to your question, feel free to post comment. Positive images Why positive images are named so? Because a positive image contains the target object which you want machine to detect. What’s vec file in OpenCV haartraining? During haartraining positive samples should have the same width and height as you define in command “-w -h size”. Is it possible to merge vec files? Yes, use Google, there are free tools, written by OpenCV’s community. I have positive images, how create vec file of positive samples? There is tool in C:\Program Files\OpenCV\apps\HaarTraining\src createsamples.cpp. createsamples -info positive_description.txt -vec samples.vec -w 20 -h 20 What’s positive description file? The matter is that, on each positive image, there can be several objects. positive_image_name num_of_objects x y width height x y width height … or No. Yes.

Related: