Vision API: Getting Started
- Andrei Dyakov
1. Subscribing to the Vision API
The Vision API is available for purchase on AWS Marketplace, where you can subscribe to different pricing plans based on your usage needs. To subscribe, you'll need an AWS account. If you don’t have one, you’ll need to sign up.
How to Subscribe via AWS Marketplace
Create or Sign In to Your AWS Account: If you don’t have an AWS account, sign up for an account here. If you already have one, sign in to your account.
Visit the Vision API Product Page: Once signed in, go to the Vision API page on AWS Marketplace.
Choose a Plan: Review the available pricing plans:
Basic Plan: Pay-as-you-go, no request limit, but a higher cost per request.
Other Plans: Includes a fixed number of requests, with additional requests charged at a lower rate than the Basic Plan.
You can choose both the plan and the subscription duration. After selecting a plan, press the Subscribe button, which will direct you to a registration form.
Fill in the Registration Form: It is important to properly fill in your personal name, email, and company name in the registration form. This information is required for proper configuration of your application and ensuring smooth integration with the API.
Submit and Wait for Confirmation: After submitting the form and receiving a confirmation message, wait a few minutes for an email with access details and further instructions. This email will be sent to the address you provided during registration.
Complete Purchase: Follow the on-screen instructions to finalize your purchase through AWS.
Pricing and Request Limits
Each pricing plan includes a fixed number of requests per contract. If your usage exceeds this quota, additional requests will incur a fee. Payable requests include:
Uploaded Images: Each image uploaded through the API counts as one request.
The Basic Plan has no limits, and you pay for each request you make, but the cost per request is higher compared to subscription plans. If you anticipate higher usage, subscription plans may offer better cost-efficiency.
2. API Reference and Postman Integration
We provide an extensive API reference that outlines all available endpoints. Whether you're analyzing interiors, detecting surfaces, or extracting metadata, this reference is your go-to guide for interacting with the API.
Postman Integration
The API reference is Postman-ready. You can import the collection into your Postman account to start making requests immediately. This is an excellent way to familiarize yourself with the API without needing to write code upfront.
Here’s how:
Click “Run in Postman”
Import the collection into your Postman workspace.
Now you’re ready to start experimenting with the endpoints!
3. API Authentication Setup
Every request to the Wizart Vision API must be authenticated using your API Token.
Upon subscribing to the API, you’ll receive an API token via email.
This token should be included in every request's headers like this:
X-Api-Key: your-api-token
4. Preparing Images for Analysis
For optimal results, ensure that the images you upload meet our recommended quality standards. Key recommendations include:
Resolution: Upload high-resolution images to improve detection accuracy.
Lighting: Ensure good lighting conditions to avoid excessive shadows or overexposed areas.
Angles: Try to capture the entire interior without extreme distortions.
Obstructions: Avoid blocking key surfaces with objects that could interfere with analysis.
For detailed requirements, refer to our Image Guidelines.
5. Building an Interactive Web App
If you're looking to create a web application that integrates Vision API with 3D visualization, we provide a step-by-step guide on how to use Three.js together with Vision API for real-time interaction.
The guide covers:
Setting up a Three.js scene
Sending images to Vision API for analysis
Applying detected surfaces and materials in 3D
Optimizing performance for a smooth experience
Read the full tutorial here: Crafting Immersive Virtual Spaces