Spyridon Thermos's repositories
generative-image-inpainting
A deep convnet for image inpainting.
MVFusionNet
MVFusionNet performs 3D object recognition and retrieval. The framework consists of two deep Neural Networks. We use the first network for deep feature extraction from rendered views of a 3D shape. In addition, a second network fuses the extracted deep features with hand-crafted features. In this repository, the two deep Neural Networks are presented. You can use your own hand-crafted features for the second network.
Sensorimotor
Deep Affordance-grounded Sensorimotor Object Recognition (CVPR 2017)
sor3d_preprocessing
Hand-object separation and depthmap colorization. Last pre-processing step for SOR3D.
cardiac_dataset_trasformer
This repository has been created to improve the young researchers' experience when using cardiac data for any medical imaging application. It consists of transformers that reorganize a DICOM-based dataset into traditional image dataset structure for better dataloader exploitation.
mechkeyboards
DIY 60% mechanical keyboard experience sharing.
spthermo.github.io
Personal website
wide-residual-networks
3.8% and 18.3% on CIFAR-10 and CIFAR-100
GaNDLF
A generalizable application framework for segmentation, regression, and classification using PyTorch
imageprocessing_course_icmc
Image Processing Course at ICMC/USP