nvidia

Efficient semantic segmentation with lightweight transformers.

Real-time object detection with competitive accuracy.

Efficiently match individuals across multiple camera views.

Classify skeleton sequences using ST-GCN models.

Train and deploy 3D object detection models from LiDAR data.

Automate defect detection using Siamese networks.

Unified image segmentation for various tasks.

Efficiently train and deploy OCRNet models for text recognition.

Efficient scene text detection in images.

Panoptic 3D scene reconstruction from RGB images.

Self-supervised visual representation learning for vision transformers.

Fine-tune models for visual recognition tasks.

Universal image segmentation with transformer-based models.

Train and deploy instance segmentation models with text prompts.

Efficiently generate segmentation masks with minimal annotations.

Efficiently pretrain and fine-tune visual models.

Streamline image classification with PyTorch-based models.

Open-set object detection with language guidance.

Estimate depth from stereo images for 3D reconstruction.

Real-time stereo depth estimation with low latency.

Train and deploy 2D object detection models with DINO.

Predict per-pixel depth from single RGB images.

Efficient 2D object detection using deformable attention.

Keypoint and object pose estimation for 6-DoF models.