nihalsid/deep-active-semantic-segmentation

Stars
2
Language
Python
Created over 5 years ago
Updated over 5 years ago

nihalsid/deep-active-semantic-segmentation

nihalsid

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Active learning for semantic segmentation with deep neural networks - documentation coming soon

mesh-gpt

MeshGPT: Generating Triangle Meshes with Decoder-Only Transformers

panoptic-lifting

[CVPR23 Highlight] Implementation for Panoptic Lifting

ViewAL

[CVPR'20] Implementation for the paper "ViewAL: Active Learning with Viewpoint Entropy for Semantic Segmentation"

retrieval-fuse

[ICCV21] Code for "RetrievalFuse: Neural 3D Scene Reconstruction with a Database"

stylegan2-ada-3d-texture

SRmeetsPS-CUDA

CUDA implementation of the paper "Depth Super-Resolution Meets Uncalibrated Photometric Stereo"

stylegan2-ada-lightning

Simplified pytorch lightning port of StyleGAN2-ADA

single-view-3d-reconstruction

texturify

Pytorch implementation for paper "Texturify: Generating Textures on 3D Shape Surfaces"

shape_sdf

Simple MLP for representing the SDF of a single shape

deepfillv2-pylightning

Clean minimal implementation of Free-Form Image Inpainting with Gated Convolutions in pytorch lightning. Inspired from pytorch implementation by @avalonstrel.

pointcloud-visualization

for rgbd -> pointcloud projection reference

deep-joint-clustering

An image clustering algorithm using neural networks

CADTextures

image-to-atlas

Project an image to the texture atlas for a given mesh

deep-learning-experiments

Small experiments on MNIST to evaluate ES and GA against SGD

simple_overfit_test

Weirdly challenging Simple MLP overfitting problem

BiGAN

Simple implementation of a BiGAN

render-browser

Quick browsing and marking of rendered images

coordconv-pytorch

Pytorch lightning implementation of CoordConv regression experiments

boring-pl-model

Simple pytorch lightning boring model

quickshift-gpu

Quickshift on cuda

stylegan2-ada-render

gamma-spot-removal-gpu

Gamma spots removal for Neutron radiography and tomography images using 'find and replace' strategy, this is a discriminative method, better than unique threshold substitution