Tagged computer-vision

Build object detection and segmentation systems with YOLO, SAM and ONNX

Apply image classification practices to CIFAR, ImageNet and similar benchmarks

Apply object detection practices for COCO, VOC and YOLO or DETR models

Train object detection, image classification and segmentation models on cloud GPUs

Run pretrained Hugging Face transformers for text, vision and speech tasks

Compute per-class precision, recall and AP for object detection results

Find weak object detection images from per-class recall, precision and AP50

Mine unique source images for object detection from embedding parquet files

Run the DEFT augmentation loop to retrain Grounding DINO object detectors

Train BEVFusion models that fuse LiDAR and camera data for 3D object detection

Train Co-DETR object detection models with hybrid assignment in TAO

Train, quantize and export Deformable DETR models for 2D object detection

Train Grounding DINO for open-set object detection driven by text prompts

Train, distill and quantize PyTorch image classifiers with many backbones

Train Mask2Former models for panoptic, instance and semantic segmentation

Train PointPillars models for 3D object detection from LiDAR point clouds

Train lightweight SegFormer models for real-time semantic segmentation

Run four parallel C-suite agent reviews and a day of autonomous business work

Count how many times an object appears in an image with computer vision

Extract video frames as image files using OpenCV

Run a full-auto pipeline from one prompt through idea, proposal, build and verify

Drive one prompt from idea through proposal, execution and verification to done

Diagnose YOLO training runs from results.csv, mAP gaps, overfitting and divergence