Skip to content
/ libblur Public

High performance blur in pure Rust using SIMD

License

BSD-3-Clause, Apache-2.0 licenses found

Licenses found

BSD-3-Clause
LICENSE.md
Apache-2.0
LICENSE-APACHE.md
Notifications You must be signed in to change notification settings

awxkee/libblur

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Fast blur algorithms library for Rust

There are some very good and blazing fast algorithms that do blurring images. Also providing convenient api for doing convolution and some edge detection filters.
Best optimized for NEON and SSE, partially AVX, partially done WASM.

You may receive gaussian blur in 100 FPS for 4K photo.

Much faster than image default blur.

When 4-channels mode is in use that always considered that alpha channel is the last.

Also there are some available options to perform blurring in linear colorspace, or if methods do not fit you f32 options also available

Performance

Most blur algorithms done very good and works at excellent speed. Where appropriate comparison with OpenCV is available. For measurement was used M3 Pro with NEON feature. On x86_84 OpenCV might be better sometimes since AVX-2 is not fully supported in library

Usage

cargo add libblur

Usage with image feature

let blurred = gaussian_blur_image(
    img,
    61,
    0.,
    EdgeMode::Clamp,
    GaussianPreciseLevel::INTEGRAL,
    ThreadingPolicy::Adaptive,
)
.unwrap();

blurred
.save_with_format("blurred.jpg", ImageFormat::Jpeg)
.unwrap();

Gaussian blur

Excellent results. Have improvements, however, much slower than any approximations slow. Use when use need gaussian methods - smoothing, anti-alias, FFT, advanced analysis etc. There are two methods of convolution, integral approximation and exact, approximation in integral form is still gaussian with 1-3% of error however about 2x faster.

Kernel size must be odd. Will panic if kernel size is not odd.

O(R) complexity.

libblur::gaussian_blur( & bytes, src_stride, & mut dst_bytes, dst_stride, width, height, kernel_size, sigma, FastBlurChannels::Channels3, GaussianPreciseLevel::EXACT);

Example comparison time for blurring image 3000x4000 RGB 8-bit in multithreaded mode with 151 kernel size.

Time(NEON) Time(AVX)
libblur(Exact) 55.89ms 119.96ms
libblur(Integral) 33.20ms 119.67ms
OpenCV 180.56ms 182.44ms

Example comparison time for blurring image 2828x4242 RGBA 8-bit in multithreaded mode with 151 kernel size.

time(NEON) Time(AVX)
libblur(Exact) 79.34ms 93.82ms
libblur(Integral) 49.10ms 91.82ms
OpenCV 177.46ms 185.30ms

Example comparison time for blurring image 3000x4000 single plane 8-bit in multithreaded mode with 151 kernel size.

time(NEON) Time(SSE/AVX)
libblur(Exact) 24.19ms 29.89ms
libblur(Integral) 13.70ms 29.34ms
OpenCV 74.73ms 64.20ms

Stack blur

The fastest with acceptable results. Result are quite close to gaussian and look good. Sometimes noticeable changes may be observed. However, if you'll use advanced analysis algorithms non gaussian methods will be detected. Not suitable for advanced analysis. Results just a little worse than in 'fast gaussian', however it's faster.

O(1) complexity.

libblur::stack_blur( & mut bytes, stride, width0, height, radius, FastBlurChannels::Channels3);

Example comparison time for blurring image 3000x4000 RGB 8-bit in multithreaded mode with 77 radius.

time(NEON) time(SSE)
libblur 7.71ms 13.60ms
OpenCV 8.43ms 9.80ms

Example comparison time for blurring image 2828x4242 RGBA 8-bit in multithreaded mode with 77 radius.

time(NEON) time(SSE)
libblur 7.41ms 11.91ms
OpenCV 8.00ms 9.62ms

Fast gaussian

Very fast. Result are quite close to gaussian and look good. Sometimes noticeable changes may be observed. However, if you'll use advanced analysis algorithms non gaussian methods will be detected. Not suitable for advanced analysis. Do not use when you need gaussian. Based on binomial filter, generally speed close, might be a little faster than stack blur ( except NEON or except non multithreaded stack blur, on NEON much faster or overcome non multithreaded stackblur ), however results better as I see. Max radius ~320 for u8, for u16 will be less.

O(log R) complexity.

libblur::fast_gaussian( & mut bytes, stride, width0, height, radius, FastBlurChannels::Channels3);

Example comparison time for blurring image 3000x4000 RGB 8-bit in multithreaded mode with 77 radius.

time(NEON) time(SSE)
libblur 8.51ms 14.91ms
OpenCV - -

Example comparison time for blurring image 2828x4242 RGBA 8-bit in multithreaded mode with 77 radius.

time(NEON) time(SSE)
libblur 8.44ms 13.39ms
OpenCV -- --

Fast gaussian next

Very fast. Produces very pleasant results close to gaussian. If 4K photo blurred in 10 ms this method will be done in 15 ms. Max radius ~150-180 for u8, for u16 will be less.

O(log R) complexity.

libblur::fast_gaussian_next( & mut bytes, stride, width, height, radius, FastBlurChannels::Channels3);

Example comparison time for blurring image 2828x4242 RGBA 8-bit in multithreaded mode with 35 radius.

time(NEON)
libblur 9.41ms
OpenCV -

Example comparison time for blurring image 3000x4000 RGB 8-bit in multithreaded mode with 77 radius.

time(NEON)
libblur 9.28ms
OpenCV -

Tent blur

2 sequential box blur ( theory ) that produces a tent filter. Medium speed, good-looking results with large radius tents becoming more noticeable

O(1) complexity.

libblur::tent_blur(bytes, stride, & mut dst_bytes, stride, width, height, radius, FastBlurChannels::Channels3);

Median blur

Median blur ( median filter ). Implementation is fast enough.

O(log R) complexity.

libblur::median_blur(bytes, stride, & mut dst_bytes, stride, width, height, radius, FastBlurChannels::Channels3);

Example comparison time for blurring image 3000x4000 RGB 8-bit in multithreaded mode with 35 radius.

time(NEON) time(SSE)
libblur 603.51ms 872.03ms
OpenCV 637.83ms 959.07ms

Example comparison time for blurring image 2828x4242 RGBA 8-bit in multithreaded mode with 35 radius.

time(NEON) time(SSE)
libblur 643.22ms 695.75ms
OpenCV 664.22ms 808.21ms

Gaussian box blur

Generally 3 sequential box blurs it is almost gaussian blur ( theory ), slow, really pleasant results. Medium speed.

O(1) complexity.

libblur::gaussian_box_blur(bytes, stride, & mut dst_bytes, stride, width, height, radius, FastBlurChannels::Channels3);

Box blur

Box blur. Compromise speed with bad looking results. Medium speed.

O(1) complexity.

libblur::box_blur(bytes, stride, & mut dst_bytes, stride, width, height, radius, FastBlurChannels::Channels3);

Example comparison time for blurring image 3000x4000 RGB 8-bit in multithreaded mode with 77 radius.

time(NEON) time(SSE)
libblur 15.52ms 23.01ms
OpenCV 16.81ms 24.38ms

Example comparison time for blurring image 2828x4242 RGBA 8-bit in multithreaded mode with 77 radius.

Time(NEON) Time(SSE)
libblur 12.79ms 24.32ms
OpenCV 16.33ms 23.75ms

Fast bilateral blur

This is fast bilateral approximation, note this behaviour significantly differs from OpenCV. This method has high convergence and will completely blur an image very fast with increasing spatial sigma. By the nature of this filter the more spatial sigma are the faster method is.

fast_bilateral_filter(
    src_bytes,
    & mut dst_bytes,
    dimensions.0,
    dimensions.1,
    kernel_size,
    spatial_sigma,
    range_sigma,
    FastBlurChannels::Channels3,
);

This project is licensed under either of

  • BSD-3-Clause License (see LICENSE)
  • Apache License, Version 2.0 (see LICENSE)

at your option.