[Kaggle] Minority Report, or the San Francisco Random Forest Precog

I had a little free time on my hand and decided to quickly complete the coursera-course „Data Science at Scale – Practical predictive analytics“ of the University of Washington by Bill Howe. The last assignment was to participate in a kaggle competition.

Source: https://commons.wikimedia.org/wiki/File:ExpoSYFY_-_Minority_Report_(10825723756).jpg

Source: https://commons.wikimedia.org/wiki/File:ExpoSYFY_-_Minority_Report_(10825723756).jpg

For this assignment I chose the „San Francisco Crime Classification“ challenge. The task is to predict the Category of a crime given the time and location. The dataset contains incidents from the SFPD Crime Incident Reporting system from 2003 to 2015 (878049 datapoints for training) with the following variables: (mehr …)

Short: Create MJPG stream to view in browser from Opencv using python

I had to google some protocols and methods of opencv, I consider it worth a short.

Goal: Do some realtime image processing of a webcam and directly stream the resulting images to a browser as mjpg.
(mehr …)

Smarter smartmirror

So I also decided to build myself a smartmirror. However, I want it to provide a little more functionality than just displaying some information and telling me that I’m beautiful. Here is the finished build:

Bathroom SmartMirror With LeapMotion


And here is a video of the leap-motion-control in action:


I want to place it in my bathroom, because that’s the only place where I actually spend some time in front of the mirror. I do want some controls, but I do not want to touch buttons or the mirror itself, so I chose a leap motion controller. Below I will detail some of the steps I went through in building this thing.

(mehr …)

Groundtruth data for Computer Vision with Blender

In the last few days I took a look at blender again, and discovered that you can use Nodes to save individual render passes.
In the video below you can see the sequence of a car driving in a city scene and braking. The layers I rendered out for groundtruth data are the rendered image with the boundingbox of the car (top left), the emission layer ( shows the brakelights when they start to emit light, top right ), the optical flow (lower left), and the depth of each pixel in the world scene ( lower right).


Render-time was about 10h on a Nvidia GeForce GTX 680, tilesize 256×256, total image-size: 960×720.


Displacement priors

What is the target of all this ? Driving in an automotive scenario with a given speed and turnrate at any moment, we want to predict the displacement of a 2D-projection (pixel) between two frames:
p(\vec{uv}_{x,y} | speed, turnrate, camera-matrix, world-geometry)

By using the camera-calibration, I can create artificial curves and walls as 3D point-sets and project them back to 2D. Using discretized values for speed, turnrate, streetwidth and wall-height, I can then simulate the displacement of these 3D-Points when they are projected to 2D (our image).
(Note for me: this is the backprojection-code, main-file: main_displacements.py)


(mehr …)

Symmetry detection

This will probably become one of our modalities in the future: symmetry !

Thanks to the guys at hs-niederrhein, there is symmetry-detection-code that can already be used
for some first estimates:

This software implements the gradient product transform for symmetry
detection that is described in the paper

C. Dalitz, R. Pohle-Froehlich, F. Schmitt, M. Jeltsch:
„The gradient product transform for symmetry detection
and blood vesselm extraction.“ International Conference on
Computer Vision Theory and Applications (VISAPP), pp. 177-184,

And the first results look quite promising:

Lane detection

Today I will try to detect some lanes..

– We know the lane-width (plus minus)
– We are inside the middle of a lane
– We know the camera geometry
– Based on the turnrate of the IMU we can estimate the curvature of the street
– A line in pixels can be detected by a upward flank and a downward flank

Here are some exemplary results:

1) Of course, the best one first 😉

(mehr …)

Training Cascades to detect cars

I spent some time on training several cascades to detect cars in ego-view automotive videos,
and will now document what I’ve learned.

I will use the existing OpenCV-tools.

Data preparation
-> pos/ 1000 images containing the desired object
-> pos.info (containing the filenames of the objects, number of objects in the frame and bounding boxes in the format x,y,width,height)
-> neg/ 2000 images that do not contain cars at all
-> negs.txt text-file containing the filenames of all negative images

For the positive images I used tight bounding boxes. You actually do not need as many negative images as you want to use negative samples later on, as the training-script will sample patches from the negative images given, so it can actually be less images than negative samples.

Some of the positive images, the bounding-boxes have been annotated by hand (ground-truth-data):





(mehr …)

BCCN 2015 Poster

We presented our poster at the BCCN conference 2015 in Heidelberg. It describes our system platform and a first case study of brakelight detection.

Download the PDF file .