Sylvester APIGuideReference

Sylvester API Overview: AI-Powered Mood Detection for Cats

User Manual v1.0

Understand Your Cat’s Emotions—With a Photo

The Sylvester.ai API is a production-ready REST API powered by Sylvester.ai’s cutting-edge machine learning technology. With a single photo, Sylvester analyzes a cat’s facial expression and returns whether the cat appears happy or not happy, complete with confidence scores and supporting metadata.

This API is designed for veterinary platforms, pet wellness apps, telehealth providers, and insurers looking to elevate feline care with smart, non-invasive emotional assessments

🔍 What the API DoesCopied!

Sylvester uses computer vision and a proprietary machine learning model to analyze images of cats and return:

  • Mood Classification: Happy or Unhappy

  • Confidence Score: Probability indicating certainty

  • Cat Detection Quality: Measure of how confidently the image contains a detectable cat

  • Facial Bounding Box & Dimensions: For further analysis or UI overlays

  • Metadata Echo: Custom values you submit for traceability (e.g. petId, customerId)

All this happens through a simple REST call with either a base64-encoded image or a URL to a hosted photo.

🎯 Who It's ForCopied!

Sylvester.ai is a perfect fit for:

  • Veterinary Clinics: Screen patients during intake, encourage follow-up visits

  • Pet Wellness Apps: Empower users to monitor emotional health at home

  • Telehealth Providers: Support remote diagnosis and triage

  • Pet Insurers: Flag patterns of pain to inform claims or coverage decisions

🚀 Key BenefitsCopied!

Benefit

Description

Non-Invasive

No physical sensors or intrusive techniques—just a photo

Veterinarian-Backed

Trained and validated against expert behaviorist data

Simple Integration

Easy REST API with clear responses, built for developers

Engaging for Pet Owners

Gives pet parents new insights into how their cats feel

Proactive Care

Encourages early vet visits before symptoms worsen

🧠 The AI Behind ItCopied!

Sylvester.ai is powered by a VGG-based convolutional neural network, trained on thousands of annotated feline images from clinics, shelters, volunteers, and real users. Preprocessing includes facial feature detection, brightness normalization, and noise filtering.

Accuracy:
On clear cases where expert behavioral scores indicate pain, the model shows up to 80% agreement with trained professionals.

⚙️ How It WorksCopied!

🖼️ Submit a Scan

Send a POST request to /scan with:

  • Image: Base64-encoded image or URL to image

  • Optional Metadata: Custom info like pet ID, customer ID, or tags

The system responds with a unique scanId.

🔄 Check Status / Get Result

Query /scan/{scanId} to retrieve the full result, including:

  • Detected mood

  • Probability/confidence scores

  • Whether a cat was found

  • Optional facial bounding box data

  • Timestamped logs for historical tracking

🧪 Sample ResponseCopied!

{
  "status": "COMPLETED",
  "scanId": "abc123",
  "isCat": true,
  "isHappy": false,
  "predictions": {
    "prediction": "Unhappy",
    "probability": 0.89,
    "cat_quality": 0.78,
    "message": "Success"
  },
  "createdAt": "2025-04-09T12:00:00Z"
}

⚠️ Limitations & Best PracticesCopied!

Factor

Recommendation

Breed

Lower accuracy for Persian, Siamese, Sphynx, etc.

Image Size

Max 4MB

Clarity

Avoid blurry images

Face Angle

Front-facing cat face only

Lighting

Avoid shadows or overly dark images

Age

Best results on adult cats

Use consistent image collection methods and test your flows in our sandbox environment to optimize results.

🧩 Integration SummaryCopied!

  • Auth: API key + secret headers

  • Methods:

    • POST /scan: Submit image for analysis

    • GET /scan/{scanId}: Retrieve result

  • Data: JSON payloads, base64 or image URLs

  • Tools: Postman collection and evaluation environment available on request

📈 Use Cases in ActionCopied!

  • Case 1: A cat owner received a “Not Happy” result. Vet diagnosed ear mites before any visible symptoms.

  • Case 2: Cat with subtle dehydration symptoms flagged via Sylvester.ai. Vet administered fluids.

  • Case 3: 16-year-old cat showed discomfort. Vet prescribed arthritis medication. Owner noted immediate behavior improvement.

🧪 Try It YourselfCopied!

We offer:

  • Access to a sandbox environment for API testing

  • Postman collections to simulate requests

  • Partner dashboard for monitoring usage and results

  • Dedicated support for integration

🔗 Technical ReferenceCopied!

Full OpenAPI/Swagger documentation is available here:
https://docs.sylvester.ai/reference

📬 Contact & Next StepsCopied!

Interested in evaluating the Sylvester.ai API or integrating it into your product?

📧 Email: hello@sylvester.ai
🌐 Website: www.sylvester.ai

We’re happy to provide demo access, technical onboarding, and guidance to ensure a seamless launch.