# Grounding Dino

> Grounding DINO zero-shot object detection — natural-language queries to labeled 2D bounding boxes with confidence scores. Use when a workflow needs to locate named objects in an RGB image before segmenting or grasping them.

- Skill: `graph-robots/grounding-dino` (Agent Skill, multi-file: 5 files)
- Install (CLI): `npx skillmds@latest add graph-robots/grounding-dino`
- Raw SKILL.md: https://api.skillmd.com/api/skills/graph-robots/grounding-dino/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Productivity
- License: MIT
- Author: graph-robots (https://skillmd.com/u/graph-robots)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/graph-robots/grounding-dino

---


# grounding-dino

The Grounding DINO servicer (`IDEA-Research/grounding-dino-base` via
transformers) as one in-process tool. Image in: RGB uint8 `[H, W, 3]` numpy
array; out: `{detections: [{box, label, score}, ...]}`.

## When to use

- Locating a named object in a camera frame: `grounding-dino.detect(rgb,
  "cream cheese box.")`, pick the best box (highest score, or closest to a
  pointing-model pixel), then `sam3.segment_box` for a pixel-accurate mask.
- Empty `detections` means nothing cleared the thresholds — treat as
  not-found, don't retry blindly with the same prompt.

## Install

```bash
uv sync --extra grounding-dino   # torch + transformers
# (pip: pip install -e ".[grounding-dino]")
```

Weights download from HuggingFace on first call. Env knobs:
`GAP_DINO_DEVICE` (default `cuda`; CPU works but is slow) and
`GAP_DINO_MODEL` (default `IDEA-Research/grounding-dino-base`).

## Gotchas (carried over from the servicer)

- **Period-separated phrases**: GDINO's text encoder expects each object
  phrase terminated with `.` (`"red cube. green cube."`). A missing final
  period is appended automatically, but separate multiple objects yourself.
- Default thresholds are deliberately low (0.20/0.20) for recall on
  household objects; raise them when false positives leak through. Zero or
  negative thresholds fall back to the defaults (proto-default semantics).
- `label` strings are the matched text spans, not your full query — when
  querying multiple phrases, group detections by label.
- The model is a lazy module-level singleton; the first call pays the
  weights-load latency, subsequent calls don't.