Back to Documentation•ML Models

Computer Vision APIAdvanced image and video analysis powered by AI

Comprehensive computer vision capabilities including image classification, object detection, face recognition, and video analysis.

Vision API

Overview

Computer Vision API

Our Computer Vision API provides state-of-the-art image and video analysis capabilities. Process images and videos to extract insights, classify objects, detect faces, and analyze visual content with industry-leading accuracy.

Base URL

https://api.litends.com/v2/vision

Supported Formats

Images: JPEG, PNG, WebP, BMP
Videos: MP4, AVI, MOV, WebM
Max file size: 50MB
Max resolution: 4K

Key Features

Object detection & classification
Face detection & recognition
Optical character recognition
Video content analysis

Image Classification

POST/v2/vision/classify

Classify objects and scenes in images

Parameters

Name	Type	Required	Description
`image_url`	string	Optional	URL of the image to classify
`image_data`	string	Optional	Base64 encoded image data
`max_results`	integer	Optional	Maximum number of classifications to return (default: 10)
`confidence_threshold`	float	Optional	Minimum confidence score (0.0-1.0, default: 0.5)

Request Example

JSON

{
  "image_url": "https://example.com/image.jpg",
  "max_results": 5,
  "confidence_threshold": 0.7
}

Response Example

JSON

{
  "classifications": [
    {
      "label": "Golden Retriever",
      "confidence": 0.92,
      "category": "animal"
    },
    {
      "label": "Dog",
      "confidence": 0.89,
      "category": "animal"
    },
    {
      "label": "Pet",
      "confidence": 0.78,
      "category": "general"
    }
  ],
  "image_properties": {
    "width": 1920,
    "height": 1080,
    "format": "jpeg"
  },
  "processing_time_ms": 125
}

POST/v2/vision/analyze

Comprehensive image analysis including objects, faces, and text

Parameters

Name	Type	Required	Description
`image_url`	string	Optional	URL of the image to analyze
`image_data`	string	Optional	Base64 encoded image data
`features`	array	Optional	Analysis features to enable: ["objects", "faces", "text", "colors"]
`include_metadata`	boolean	Optional	Include detailed metadata in response

Request Example

JSON

{
  "image_data": "data:image/jpeg;base64,/9j/4AAQSkZJRgABAQEAYABgAAD...",
  "features": ["objects", "faces", "text"],
  "include_metadata": true
}

Response Example

JSON

{
  "objects": [
    {
      "name": "person",
      "confidence": 0.94,
      "bounding_box": {
        "x": 120, "y": 50,
        "width": 200, "height": 300
      }
    }
  ],
  "faces": [
    {
      "confidence": 0.98,
      "bounding_box": {
        "x": 150, "y": 75,
        "width": 80, "height": 100
      },
      "emotions": {
        "happiness": 0.85,
        "surprise": 0.12
      }
    }
  ],
  "text": [
    {
      "text": "Welcome to AI",
      "confidence": 0.96,
      "bounding_box": {
        "x": 50, "y": 400,
        "width": 300, "height": 40
      }
    }
  ],
  "processing_time_ms": 189
}

Object Detection

POST/v2/vision/detect

Detect and locate objects in images

Parameters

Name	Type	Required	Description
`image_url`	string	Optional	URL of the image to process
`image_data`	string	Optional	Base64 encoded image data
`object_types`	array	Optional	Specific object types to detect
`min_confidence`	float	Optional	Minimum detection confidence (default: 0.6)

Request Example

JSON

{
  "image_url": "https://example.com/street-scene.jpg",
  "object_types": ["person", "car", "bicycle", "traffic_light"],
  "min_confidence": 0.7
}

Response Example

JSON

{
  "detections": [
    {
      "object": "person",
      "confidence": 0.94,
      "bounding_box": {
        "x": 245,
        "y": 120,
        "width": 85,
        "height": 180
      }
    },
    {
      "object": "car",
      "confidence": 0.89,
      "bounding_box": {
        "x": 450,
        "y": 200,
        "width": 200,
        "height": 120
      }
    }
  ],
  "total_objects": 2,
  "processing_time_ms": 156
}

Video Analysis

POST/v2/vision/video/analyze

Analyze video content for objects, activities, and scenes

Parameters

Name	Type	Required	Description
`video_url`	string	Optional	URL of the video to analyze
`video_data`	string	Optional	Base64 encoded video data
`frame_interval`	integer	Optional	Analyze every Nth frame (default: 30)
`features`	array	Optional	Analysis features: ["objects", "activities", "scenes"]

Request Example

JSON

{
  "video_url": "https://example.com/video.mp4",
  "frame_interval": 15,
  "features": ["objects", "activities"]
}

Response Example

JSON

{
  "video_info": {
    "duration": 120.5,
    "fps": 30,
    "resolution": "1920x1080",
    "total_frames": 3615
  },
  "timeline": [
    {
      "timestamp": 0.5,
      "objects": ["person", "bicycle"],
      "activities": ["cycling"],
      "confidence": 0.91
    },
    {
      "timestamp": 1.0,
      "objects": ["person", "bicycle", "car"],
      "activities": ["cycling", "driving"],
      "confidence": 0.88
    }
  ],
  "summary": {
    "dominant_objects": ["person", "car", "bicycle"],
    "main_activities": ["cycling", "walking", "driving"],
    "scene_type": "urban_street"
  },
  "processing_time_ms": 4500
}

Ready to add computer vision to your app?

Start processing images and videos with our powerful computer vision APIs. From simple classification to complex video analysis.

Python SDK

View Tutorials