object-detection

Name: object-detection
Availability: InStock
Author: JigsawStack

Detect objects in images and PDFs with pixel-space bounding box coordinates.

4|Updated Mar 10, 2026

Search Tags:#computer vision #object detection #image understanding #bounding boxes #OCR localization #spatial coordinates #pdf vision

Ready to run

npx skills add https://github.com/JigsawStack/interfaze-skills --skill object-detection

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: object-detection
Source: https://github.com/JigsawStack/interfaze-skills/tree/main/skills/object-detection
Command: npx skills add https://github.com/JigsawStack/interfaze-skills --skill object-detection

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Object detection turns an image or PDF into actionable spatial understanding by identifying items and returning their exact positions.

Core Features & Use Cases

Bounding Boxes for Detected Objects: Locate objects with pixel-space coordinates for top-left and bottom-right corners.
Optional Text Localization: Return visible text with coordinates alongside objects when requested.
Flexible Triggers: Works even when users ask indirectly (e.g., “where is the logo?”) instead of explicitly saying “object detection.”

Quick Start

Ask the AI to “Detect the crane in this image and return its bounding box coordinates, plus any visible text with positions.”

Frequently Asked Questions about object-detection

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I get bounding box coordinates for objects in an image?▼

You can locate objects by detecting them in images and returning pixel-space bounding box coordinates. The output provides top-left and bottom-right corner positions for each detected item, enabling exact spatial mapping within the visual document.

Can I extract text with spatial coordinates from a PDF?▼

Text localization is supported alongside object detection in PDFs. The system returns visible text with pixel-space bounding box coordinates, allowing you to map exactly where text appears within the visual document.

Does object detection work if I ask where a specific logo is without explicitly requesting bounding boxes?▼

Yes, object detection works with flexible triggers, functioning even when users ask indirectly. Asking "where is the logo?" returns the logo's bounding box coordinates without needing to explicitly mention object detection or spatial mapping.

What is the best way to count and locate specific targets like vehicles in a visual document?▼

Object detection locates specific targets like vehicles by producing pixel-space bounding box coordinates for each item in an image or PDF. This enables counting objects and mapping exactly what appears where in the visual document.

How do I detect a specific object and return its bounding box coordinates?▼

Provide an image or PDF to the AI and request it to detect the target object and return its bounding box coordinates. The system uses the Interfaze model endpoint to deliver consistent structured results with top_left and bottom-right coordinate fields.