
Auto-labels image-caption pairs with referring expressions and pixel bounding boxes
- Open source
- 4K GitHub stars
- Design and media

Auto-labels image-caption pairs with referring expressions and pixel bounding boxes

Generates synthetic object-detection defect images with COCO labels via AnomalyGenNext

Generates referring-expression and grounding annotations from images with KITTI bounding

Turns raw videos into chain-of-thought QA training data using VLM and LLM teacher models

Mines look-alike source images for VCN augmentation via DEFT embedding

Mines unique source images closest to target samples for object detection datasets

Runs an iterative RT-DETR retraining loop for binary industrial defect detection

Guides an agent through training, evaluating and running BEVFusion LiDAR-camera 3D

Runs and fine-tunes NVIDIA's FastFoundationStereo model for real-time stereo depth

Guides an agent through training, evaluating and running FoundationStereo stereo depth

Trains, evaluates, distills and quantizes PyTorch image classifiers inside TAO containers

Trains, evaluates, exports and runs NVPanoptix3D models for 3D panoptic scene segmentation

Trains, evaluates and exports PointPillars 3D LiDAR object detection models with TAO

Composes and deploys an NVIDIA VSS vision stack with captioning, detection, search

Calibrates multi-camera MP4 recordings through the AutoMagicCalib REST API

Guides an agent through NVIDIA's CDF AnomalyGen pipeline for synthetic defect images

Guides an agent through authoring and running Cosmos Data Factory video and image

Finds similar driving clips or generates rain, snow and daylight variations of them

Guides an agent through NVIDIA Cosmos Curator video and image dataset pipelines

Generates synthetic PCB inspection images with defects on Kubernetes via a DAG

Runs an NVIDIA pipeline that turns seed images into anomaly video clips and datasets

Runs NVIDIA's image attribute augmentation pipeline on Kubernetes via Airflow

Runs NVIDIA's video data augmentation Airflow workflow to create and label Cosmos video

Builds ready-to-run DeepStream GStreamer pipelines for video and image inference

Lets an agent set up DOCA Rivermax receive streams for timing-precise media over IP

Runs DOCA UROM Service on BlueField Arm to execute offloaded memory operations

Guides an agent through adding a new modality to NVIDIA FastGen across data, model

Guides an agent through Azure OpenAI .NET SDK code for chat, embeddings, images and audio

Guides building real-time speech-to-speech voice assistants in .NET with Azure AI Voice

Lets an agent check text and images for harmful content using Azure AI Content Safety