Train Scene Graph Generation for Visual Genome and GQA in PyTorch >= 1.2 with improved zero and few-shot generalization.

Generative modeling of synthetic time series data and time series augmentations

Custom image data generator for TF Keras that supports the modern augmentation module albumentations

Deep Illuminator is a data augmentation tool designed for image relighting. It can be used to easily and efficiently generate a wide range of illumination variants of a single image.

Repo for the paper "Extrapolating from a Single Image to a Thousand Classes using Distillation"

⚡ Blazing fast audio augmentation in Python, powered by GPU for high-efficiency processing in machine learning and audio analysis tasks.

RuTransform: python framework for adversarial attacks and text data augmentation for Russian

This repo contains full solution for the challenge: from datasets creation to training and creating submit file. Moreover it can be used as a universal high-quality baseline solution for any segmentation task.

Official implementation for "Pure Noise to the Rescue of Insufficient Data: Improving Imbalanced Classification by Training on Random Noise Images" https://arxiv.org/abs/2112.08810

Runner Up Solution for the NeurIPS 2020 Competition - "Predicting Generalization in Deep Learning"

Source code of top 3% solution for the Kaggle APTOS 2019 Blindness Detection challenge.

Repository for the paper "Exploring Image Augmentations for Siamese Representation Learning with Chest X-Rays"

Aug Tool is a Python library available on PyPI that simplifies image data augmentation for machine learning tasks, compatible with TensorFlow, PyTorch, and the YOLO library.

Augmentation yolo format txt and images

State-of-the-art https://arxiv.org/abs/2302.09119 https://intranet.matematicas.uady.mx/journal/index.php?c=50

VAE learned on augmented data to improve generalization.

2D facial landmarks detection with neural networks

Explore and build ImgAug augmentations with Supervisely

UI to prepare training data for SmartTool inside Supervisely platform