Trailer for my first game, Sack of Bots, which is now online on steam. It is a top down, score-chasing arcade shooter. Give it a look if you like : https://t.co/On1zGhnfFH
I used some of the equipment from my job
@SenseGlove
to play Jenga from a second-person point of view. It was interesting to see how quickly your motor skills can adapt to a new perspective.
#computervision#virtualreality#mixedreality
@martinozanzi Thanks! It is not on github, but it is a retrained mobilenet ssd for trafficlight detection, followed by a simple hsv filter to detect red/orange/green lights. Hope that helps!
Been working on a traffic light detection model, which can detect the correct traffic light and whether it is green/yellow/red.
Pretty happy with how it turned out.
#MachineLearning#tensorflow#opencv
I made a script that takes a paragraph of text describing a #computervision model and turns it into python code. Was fun as a thought experiment in how to make #MachineLearning more accessible. #NLP
Never thought I would get to learn 3D modelling software when I started my work @SenseGlove , but I'm glad I can because it is good fun.
I can recommend #blender for those #computervision and #machinelearning engineers looking to supplement their dataset with artificial data
Racing with SenseGlove! Our Computer Vision expert @FolmerMartijn always finds the most creative ways to improve #CV detection.
Curious to see how you can use SenseGlove for #VR training? https://t.co/sYDwCtfHx3
#computervision#motiontracking#training
Recreated one of my old projects, using Evolutionary Neural Networks to make a Flappy Bird clone learn to play itself.
Normally I just import python libraries for #MachineLearning, but recreating functionality in #GameMaker is frustrating, but very satisfying once it works βοΈ
Our Computer Vision expert @FolmerMartijn wishes you a Merry Christmas by using his custom algorithm to detect sign language.
If you are missing #touch and #motion tracking in your #VR projects, you know where to find us! https://t.co/OeggFmfrh3
With Love,
SenseGlove
@HigginsThird First couple of test rounds, I just wrote down random numbers and equations, saw the resulting graph with wild curves and said "That looks cool, but I have no idea if this makes any sense"
@BartronPolygon Yes, well spotted. I use contour detection to get the individual symbols, then assign them to an equation based on their vertical location within the frame. After that, I pass each symbol through a CNN for classification
I wanted to be able to quickly prototype an idea (such as a racetrack for a video game) by drawing it out on a piece of paper and using computer vision to translate it to something I can try out. Pretty satisfied with how it turned out.
#AugmentedReality#Computervision#gamedev
Augmented Reality : real-time recognition of handwritten math functions and drawing their graphs
Pretty satisfied with how my first prototype turned out. Next step, sin/cos/tan and integrals (and ignoring my hand).
#machinelearning#computervision#augmentedreality
For a work assignment I had to create a script which uses a neural network to detect sums on a piece of (rotated) paper and solve them. I'm pretty satisfied how it turned out
#Opencv#TensorFlow#machinelearning
@CesarOttani@GMspotlight But seriously, in the case of GM, it was a matter of combining weights and biases in well performing neural networks to create better ones, using many for loops. It was very inefficient as compared to using python (and I imagine C#), but workable