adirik/grounding-dino 🖼️📝🔢✓ → ❓

▶️ 18.1M runs 📅 Oct 2023 ⚙️ Cog 0.8.3 🔗 GitHub 📄 Paper ⚖️ License

image-object-detection open-vocabulary-detection

About

Detect everything with language!

Example Output

Output

{"detections":[{"bbox":[19,204,408,563],"label":"pink mug","confidence":0.8077122569084167},{"bbox":[545,263,952,650],"label":"pink mug","confidence":0.7644544839859009},{"bbox":[416,60,764,380],"label":"pink mug","confidence":0.4754282832145691},{"bbox":[909,161,1078,487],"label":"pink mug","confidence":0.43150201439857483}],"result_image":"https://pbxt.replicate.delivery/oJDgostFveUPEiUVnTRM9MnL7rxEMXIuy65E6K2X638PfFyRA/result.png"}

Performance Metrics

1.68s Prediction Time

1.67s Total Time

All Input Parameters

{
  "image": "https://replicate.delivery/pbxt/JlgUQIQCDemKg7bnfn5zKMqLgAPrZdpMMHzkXgHX5HUlbw9z/mugs.webp",
  "query": "pink mug",
  "box_threshold": 0.2,
  "text_threshold": 0.2,
  "show_visualisation": true
}

Input Parameters

image Type: string: Input image to query
query Type: string: Comma seperated names of the objects to be detected in the image
box_threshold Type: numberDefault: 0.25Range: 0 - 1: Confidence level for object detection
text_threshold Type: numberDefault: 0.25Range: 0 - 1: Confidence level for object detection
show_visualisation Type: booleanDefault: true: Draw and visualize bounding boxes on the image

Output Schema

Output

Version Details

Version ID: efd10a8ddc57ea28773327e881ce95e20cc1d734c589f7dd01d2036921ed78aa
Version Created: October 23, 2023

Run on Replicate →