Lightly ORG

Curate, Annotate, and Manage Your Data in LightlyStudio.

All-in-one training for vision models (YOLO, ViTs, RT-DETR, DINOv3): pretraining, fine-tuning, distillation.

A python library for self-supervised learning on images.

A tool for converting computer vision label formats.

Plugins for LightlyStudio.

Easily get basic insights about your ML dataset

uses Lightly and Label Studio to performa a complete AL workflow from dataset scraping to usage

Tutorial on active learning with the Nvidia Transfer Learning Toolkit (TLT).

Curated List of papers on Self-Supervised Representation Learning

Tutorial on active learning with the Nvidia Train, Adapt, and Optimize (TAO) toolkit.

No description.

Some images showing different kinds of cloths. One directory per kind of clothes.

The dataset includes 8479 images of 6 different fruits (Apple, Grapes, Pineapple, Orange, Banana, and Watermelon). Fruits are annotated in YOLOv8 format.

No description.

Factory visual inspection with LightlyStudio + Gemini autolabeling + LightlyTrain — notebook and video script

SAM3 text-prompted autolabeling with LightlyStudio — notebook, blog post, and video script

No description.

No description.

First 128 train and val images from the COCO dataset in the YOLO format

No description.

Example datasets for Project Purple

Ultralytics YOLO11 🚀

Repo for organization utility scripts executed as gists

RF-DETR is a real-time object detection model architecture developed by Roboflow, SOTA on COCO & designed for fine-tuning.

The largest collection of PyTorch image encoders / backbones. Including train, eval, inference, export scripts, and pretrained weights -- ResNet, ResNeXT, EfficientNet, NFNet, Vision Transformer (ViT), MobileNetV4, MobileNet-V3 & V2, RegNet, DPN, CSPNet, Swin Transformer, MaxViT, CoAtNet, ConvNeXt, and more

YOLOv12: Attention-Centric Real-Time Object Detectors

[Incl. GenAD, CVPR 2024 Highlight] Embracing Foundation Models into Autonomous Agent and System

A self-contained Jupyter notebook for running the whole Lightly solution including the installation.

Preview of a Lightly Insights HTML report

Create Lightly prediction files with custom prediction models

No description.

Example of folder structure of a complete run on DeepSportRadar Instance Segmentation Challenge dataset