nfnt
0f9f918da3
Use RGBA, RGBA64 image types as output.
...
These image types use premultiplied alpha values which are also used during the interpolation. If we'd use NRGBA, NRGBA64 as output, we'd have to reverse the premultiplication.
2016-01-07 21:30:54 +01:00
nfnt
579058ccc0
Test for color-casts due to wrong alpha-channel scaling.
2016-01-07 20:34:21 +01:00
minodisk
87f723015c
Test same color with some color modes
2015-05-25 18:53:05 +09:00
minodisk
f430a41192
Fix wrong color with RGBA input
2015-05-25 18:20:03 +09:00
nfnt
f2d1b73023
Merge branch 'alpha_channel_fix'
...
Colors in image.RGBA, image.RGBA64 and generic types with alpha channel should
now resize correctly. There's a small performance hit for these image types even
if they don't use alpha channels though.
Fixes bug #26 .
2015-05-12 16:00:13 +02:00
nfnt
563154816f
Remove debugging directives.
2015-05-12 15:24:41 +02:00
nfnt
5a6676c19e
Add test case for input data with pre-multiplied-alpha.
2015-05-12 15:18:07 +02:00
jst
618f7ff07a
Add test case for correct pixel coordinates after resize.
2015-03-27 19:48:18 +01:00
Charlie Vieth
b09281bd36
Update benchmarks to focus on conversion performance.
...
Benchmark each interpolation function for RGBA and YCbCr images.
Reduce the image size used when benchmarking to reduce the influence of
memory performance on results.
2015-03-15 17:40:38 -04:00
jst
9485f5475a
Return input image if output dimensions equal input dimensions.
2014-12-17 11:10:53 +01:00
jst
20c6afea04
Remove unnecessary benchmark
2014-09-15 11:58:05 +02:00
jst
220cad3343
Add simple test case for input image with non-trivial bounds.
2014-08-13 20:12:18 +02:00
jst
016a61cd31
Optimize data-locality for a huge increase in processing speed.
...
This is a complete rewrite! The tight scaling loop needs data locality for optimal performance. The old version used lots of pointer redirections to access image data which was bad for data locality. By providing the complete loop for each image type, this problem is solved. Unfortunately this increases code duplication but the result should be worth it: I could measure a ~6x speed-up for certain test cases!
2014-07-19 13:19:31 +02:00
jst
446eea2f7b
Remove LUT based filters.
...
Caching kernel weights makes using the LUT based approach obsolete. Now they seem to be even slower than their counterparts.
2014-01-28 18:54:57 +01:00
jst
8f586c4f06
Add benchmarks to compare execution times.
...
Run "go test -bench LargeJpegThumb" to benchmark all functions at once.
2014-01-18 11:55:34 +01:00
jst
252ccc84a2
Add benchmarks for Lanczos3Lut to compare it with Lanczos3 performance.
2013-07-08 20:55:46 +02:00
jst
da95f74904
go fmt
2013-03-10 11:55:50 +01:00
jst
852f4a47e7
Merge branch 'no-image-shift' of http://github.com/jsummers/nfnt.resize
...
Merge branch 'fix-canvas-size-issue'
2013-03-10 11:33:44 +01:00
jst
84039c403b
Test for correct canvas size.
2013-03-10 10:58:08 +01:00
Jason Summers
3d9094eb9e
Center the resized image on the canvas, instead of shifting it slightly
2012-11-27 20:38:19 -05:00
jst
3e06045c3f
Speed up computation: Try to avoid Image.At() as much as possible -> specialized color access for some image types
2012-09-21 20:02:25 +02:00
jst
e548f52385
Blur input image during downscaling by scaling the filter kernel to prevent moires in the output image
2012-09-19 21:03:56 +02:00
jst
d0b2b9bc39
Added MitchellNetravali, changed Lanczos a bit
2012-09-19 19:32:00 +02:00
jst
3fc31c95cc
Kernel simplified
2012-09-15 19:30:32 +02:00
jst
339b8fd43a
Speed up calculation by avoiding dynamic casting
2012-09-01 00:21:10 +02:00
jst
9041d4db18
Function signature changed again, no need for multiple return value
2012-08-09 18:58:57 +02:00
jst
14d51db8b4
Changed function signature to include error handling. Filters simplified.
2012-08-08 21:32:51 +02:00
nfnt
fdc4a64918
Tests added
2012-08-03 18:12:26 +02:00
nfnt
428642c9f1
initial commit
2012-08-02 21:59:40 +02:00
