Cat Face Recognition Analysis

Problem

In multi-cat households, pet boarding, or home surveillance, saying this cat is often not enough: you need to know which specific individual appears in a photo or video, for feeding, monitoring, or behavior review. This skill works on cat face images or videos. After a cat profile database is registered, it compares new material against that database and returns a structured analysis report, with cloud-based historical report lookup.

How It Works

It breaks the task into four steps: cat face detection, feature extraction, database comparison, and individual recognition. Inputs can be local files or public URLs:
- --input: local image or video path
- --url: public URL, downloaded by the API service
- --list: list historical cat face recognition reports
- --detail: output level: basic, standard, or json
- --output: optional result file path

The script runs in the skill root via scripts.cat_face_recognition_analysis. User identity is handled internally and should not be requested, displayed, or exposed. Historical reports are fetched from the cloud API and rendered as a Markdown table, rather than being rebuilt from local memory or ad hoc summaries.

Boundaries and Notes

Accuracy depends on registration quality and capture conditions: the cat face should be frontal, complete, clear, and well lit, with minimal occlusion or motion blur. Supported formats include jpg/jpeg/png/mp4/avi/mov, up to 10MB. Results can be affected by angle, lighting, and occlusion, so treat them as auxiliary judgment rather than absolute identification. For historical reports, prefer the latest complete data returned by the cloud interface over local records.