API Documentation
  • Welcome!
  • Quick Start
  • API methods
  • Easy integration
  • Handling Large File Sets
  • Supported formats
  • Webhooks
  • Common usecases
    • Auto enhance image quality
    • Create business photo or avatar from face image
    • Face swap
    • Create beautiful product photo
    • Genarate image in high resolution
    • Remove background
    • AI Drawing to Image - Doodle
    • Real estate
    • Enhancing documents
    • Car dealer photo
  • Image processing
    • Resize and padding
    • Denoise and sharpen
    • Enhance lighting and colors
    • Enhance face details
    • Background removal and generation
    • Image generation
    • Inpainting and outpainting (uncrop)
    • Frame identification
    • Print
    • Captions
    • Additional parameters
    • Presets
  • Account & settings
    • Account information
  • storages
    • Description
    • AWS S3
    • AWS S3 IAM Configuration
  • Presets
    • E-commerce
    • Real-estate
    • Print/Photo
    • Digital Art
  • Technology
    • Algorithms
    • Upscale
    • Background removal
      • Remove BG recommendation
    • Sharpen & Noise reduction
    • Enhance Lighting
  • FAQ
Powered by GitBook
On this page
  1. Technology

Algorithms

PreviousDigital ArtNextUpscale

Last updated 2 years ago

Thanks to Artificial Intelligence and Machine Learning, Deep-Image.ai is trained to provide desired results in various business scenarios. We can teach algorithm-specific use cases based on a client's needs.

Deep-Image.ai uses its GPU hardware infrastructure to deliver results to clients in Europe and America. We have enough capacity to process hundreds of thousands of images monthly. We prepare a dedicated infrastructure for larger volumes of transformations; the service can also be hosted in cloud environments.

Working with neural networks does not look like a standard programming process. More like science, based on showing the pattern. Check this:

  • in Deep-Image.ai on the input, we have low-resolution graphics, while on the output we have to get a high resolution. At the beginning of cooperation with the neural network, we set random parameters

  • From the moment of entry, the neural network learns how to create good-quality graphics with the help of various transformations

  • The network counts an error by analyzing the difference between the input (the starting image) and the output (the final image). Then it modifies the weights so that the difference between successive exits is as small as possible.

  • The learning and creation process is based on algorithms (filter sets). Neural network assimilates information about a given edge so that in the end the line is smooth.

Application operation is based on the iterative process, which brings the final graphics almost to perfection.