https://github.com/mortafix/quickshift

A working implementation of Quickshift algorithm in CUDA, GPU-compatible.
https://github.com/mortafix/quickshift

cuda gpu-computing quickshift

Last synced: about 1 month ago
JSON representation

A working implementation of Quickshift algorithm in CUDA, GPU-compatible.

• Host: GitHub
• URL: https://github.com/mortafix/quickshift
• Owner: Mortafix
• Created: 2020-07-20T13:59:42.000Z (over 4 years ago)
• Default Branch: master
• Last Pushed: 2020-07-29T10:12:55.000Z (over 4 years ago)
• Last Synced: 2024-11-14T00:33:57.528Z (3 months ago)
• Topics: cuda, gpu-computing, quickshift
• Language: C
• Homepage:
• Size: 138 KB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 1
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Quickshift

A working implementation of Quickshift algorithm in CUDA, GPU-compatible.

## Usage

`Quickshift     `

* `input`: input image with a compatible formats `JPG`, `PNG`, `PNM` or `BMP`.

* `mode`: computing mode, `GPU` or `CPU`. (deafult `GPU`)

* `sigma`: approximation value to control the scale of the local density. (integer)

* `alpha`: max distance level in the hierarchical segmentation that is produced. (integer)

* `texture`: adding texture memory support, only for `GPU` mode, `y` or `n`. (default `n`)

## Examples

`Quickshift rally.jpg gpu 5 8 y`  

GPU algorithm with texture memory, JPG image as input, density 5 and distance 8.  

  

`Quickshift venice.png cpu 3 7`  

CPU algorithm, PNG image as input, density 3 and distance 7.