Louis Foucard

w-net. w-net: a convolutional neural network architecture for the self-supervised learning of depthmap from pairs of stereo images.

DepthMap_dataset. Python Blender script to generate random scene, depth map and corresponding Stereo images for machine learning.

StereoConvNet. Stereo convolutional neural network for depth map prediction from stereo images

DSB17_3d_lung_nodule_classifier. 3d convnet for the classification of nodules/tumor in lung CT scans. Trained on Luna16 for Kaggle's 2017 data science bowl competition (result in top 3%)

MC_DCNN. Multi channel deep convolutional neural network for time series classification

gym-blender. Blender Environment for Gym (work in progress)

3D_scanner. blender python script to receive and process data from an arduino/lidar/mpu6050 based 3D_scanner

wikipedia_classifier. wikipedia category scrapper, and SVM classifier

low_cost_robot_vr_control. add a vr controller (cheaper than building leader arm), flexible

draper_sat_conv_lstm. Jupyter Notebook

cuda_kernel_tensorflow. cuda implementation of a matmul operation with shared memory, as a tensorflow kernel -for reference.

RemotePresence. Remote presence with the Oculus Rift DK1 headset, Arduino, stereo webcams and a python interface.

Virtual_car. This project proposes to use a convolutional Neural network to control a Self-driving car in virtual environment generated by the Blender Game Engine.