Name: Pedro Herruzo
Type: User
Company: UNISON
Bio: Deep Learning * Spatiotemporal Data * 3D Perception & Reconstruction * VR/AR. Prev: Weather & Traffic Forecasting * Data Fusion * Generalization.
Twitter: HerruzoPedro
Location: San José, California.
Blog: https://pherrusa7.github.io/
Pedro Herruzo's Projects
Funny game to get start with Unity3D
Some caffe (http://caffe.berkeleyvision.org/) tricks and utilities to share
General Assembly's Data Science course in Washington, DC
Examples code
A convolutional Neural Network which recognize food classes by pictures
A Simple pytorch implementation of GradCAM and GradCAM++
Find interest Points in your city for your customers under some criteria
Source code for the LabelMe annotation tool.
ML and DL related contests, competitions and conference challenges.
A list of public machine learning/data science/AI contests.
Monocular odometry using OpenCV
sample code for our traffic4cast NeurIPS 2019 competition
Resources (papers, datasets, rendering methods) in the domain of object pose estimation.
Orkid Media Engine (C++/Lua/Python3/Linux/MacOs/OpenVR)
A comprehensive 10-page probability cheatsheet that covers a semester's worth of introduction to probability.
Filmmaking toolkit for depth sensors and high resolution color cameras
Some scripts that help you downloading and processing satellite data from L2 to L3 products form Sentinel-5P mission: "perform atmospheric measurements with high spatio-temporal resolution, to be used for air quality, ozone & UV radiation, and climate monitoring & forecasting." Note that L2 products are processed measurements provided by ESA (https://sentinels.copernicus.eu/web/sentinel/missions/sentinel-5p) and L3 is the projection of these measurements into a common grid or raster (https://desktop.arcgis.com/en/arcmap/10.3/manage-data/raster-and-images/what-is-raster-data.htm)
I did research in all kind of visual explanations for CNNs since saliency maps and another techniques ending up with Grad-CAM, the newer, which I have implemented a general version that can be used for any model in keras.
Code accompanying our IARAI Weather4cast Challenge
Starter Kit for the NeurIPS 2022 Weather4cast competition
Code accompanying our IARAI Weather4cast Challenge