Name: Juan Felipe Jaramillo
Type: User
Company: Universidad Politécnica de Valencia
Bio: Building AI for the future of IoT! M.Sc. AI + Electronic Engineer. Focused on resource-efficient computer vision.
Location: España
Juan Felipe Jaramillo's Projects
Step-by-step notebook for enhancing the angular resolution of light fields using a deep learning-based network, improving the resolution of the captured scene's depth map.
Labs for Coursera Applied Kubernetes
Página Web del curso de Machine learning
Depth Anything V2. A More Capable Foundation Model for Monocular Depth Estimation
Development of an Embedded System for Depth Map Estimation and Visualization using a Microcontroller with 8-bit Integer Quantization, a Raspberry Pi Server and WebSocket Protocol.
Implementation of the novel method DOD for object detection with depth estimation integrated within the prediction heads of a fully state-of-the-art convolutional neural network of one million of parameters.
Repo for pytest, containers, cloud enviorements and more
Depth Object Detector implemented on Container for CI/CD (Continuous Integration and Continuous Delivery)
Implementation of a convolutional neural network for face detection in images, achieving high accuracy (99.5% F1-score) on a diverse dataset, with an efficient inference pipeline and script for practical usage.
Classification benchmark with different architectures applied for the Geo Fossils-I dataset
Build and test my first Rust project
See: https://jaramilloh.github.io/GPC-3D-figures/ for 3D models programmed with https://threejs.org/
Implementation of an encoder-decoder convolutional neural network based on state-of-the-art models SegFormer and MA-Net for semantic segmentation in autonomous driving task.
My personal repository.
Forked from nogibjj/coursera-applied-de-kubernetes-lab
mlops template
Open Source Computer Vision Library
Algoritmos para el post-procesamiento de un campo de luz capturado; extracción de imágenes de sub-apertura de un campo de luz capturado con una cámara plenóptica afocal, sub-muestreo espacial y angular del campo de luz y secuencia sin discontinuidades utilizando el flujo óptico.
Extract sub-aperture images from captured light fields with Lytro Camera. Moreover, calibration data is given in order to rectify and color correct each capture before processing.
Red neuronal para sub-muestrear espacialmente imágenes. Proyecto realizado para la asignatura de Inteligencia Artificial en la Pontifica Universidad Javeriana.
MLOps Deploy Solutions with Rust
This repo will contain source code and materials for the TecoGAN project, i.e. code for a TEmporally COherent GAN