Popular GitHub Repositories for Randomization

Perception toolkit for sim2real training and validation in Unity

NVIDIA Deep learning Dataset Synthesizer (NDDS)

Deep Reinforcement Learning for Robotic Grasping from Octrees

SynthDet - An end-to-end object detection pipeline using synthetic data

Always sparse. Never dense. But never say never. A Sparse Training repository for the Adaptive Sparse Connectivity concept and its algorithmic instantiation, i.e. Sparse Evolutionary Training, to boost Deep Learning scalability on various aspects (e.g. memory and computational time efficiency, representation and generalization power).

Use Unity to generate synthetic images for deep learning image segmentation in PyTorch and fastai

[WACV 2024] AnyStar: Domain randomized universal star-convex 3D instance segmentation

Syclops is a tool for creating synthetic data from 3D virtual environments with photorealistic renderings and pixel-perfect annotations.

[NeurIPS 2022] Random Normalization Aggregation for Adversarial Defense

A synthetic data generator for retail object detection, which is built with Blender and Python.

Synthetic dataset generation for vision-based prosthetic grasping

State Representations as Incentives for Reinforcement Learning Agents: A Sim2Real Analysis on Robotic Grasping

Using Blender and Python to generate synthetic data for training an 8-class YOLOv5 detector in a highway driving scenario (Sedan, SUV, Van, Truck, Bus, Warning Sign, Speed Limit Sign, Speed Cam Sign).

Simple KES blends Karras and Exponential noise scheduling for high-quality images with adaptable detail, real-time config reloading, and parameter randomization for creative flexibility.

"? C++ maze generator & solver ? Uses DFS for generation & BFS for pathfinding ? Features ASCII visualization ? with start (S) & exit (E) markers ? Randomized mazes ? Optimal path highlighting ✨ Fast ⚡ Customizable ⚙️ & extendable ? Perfect for learning algorithms ? & data structures ?️"

Terminal-based, ai-enhanced companion that crafts themed names, phrases, and poetic forms using LLMs' express synthesis, randomization, and markdown pattern engineering.

Code for Adversarial Robustness through Randomization

Chess960v2 is an innovative approach to Fischer Random Chess with double randomization of starting positions. You can run tournaments using the chess engine with different starting positions, exploring new strategies and tactics in the face of complete unpredictability in the opening.

Train your eyes. Read faster.

Fully configurable Blender-based pipeline for synthetic data generation using end-to-end domain randomization for object detection tasks.