summaryrefslogtreecommitdiffstats
path: root/tests/ref/fate/j2k-dwt
Commit message (Collapse)AuthorAgeFilesLines
* avcodec/jpeg2000dwt: Print 1 digit less in the 9/7f DWT testMichael Niedermayer2015-06-221-20/+20
| | | | | | | This avoids test failure due to differing rounding between 32 and 64bit x86 Found-by: jamrial Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* avcodec/jpeg2000dwt: use 32x32->64 multiplies in the 9/7i DWTMichael Niedermayer2015-06-221-20/+20
| | | | | | | This significantly improves the quality when the integer 9/7 transform is used Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* avcodec/jpeg2000: Move H band scaling from wavelet into quantization codeMichael Niedermayer2015-06-221-20/+20
| | | | | | | This reduces the number of operations done and is equivalent except for rounding Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* avcodec/jpeg2000: Move gainb handling into the quantization codeMichael Niedermayer2015-06-221-20/+20
| | | | | | | | thats how the specification defines it, this also improves numerical accuracy of the integer wavelet implementation. It otherwise should be equivalent, in case of overflows this can be reverted. Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* avcodec/jpeg2000dwt: Fix value of F_LFTG_X and I_LFTG_XMichael Niedermayer2015-06-211-40/+40
| | | | | | | | Even if the jpeg2000 spec uses a wrong value this does not make mathematics work this way, also this has been corrected in the 2004 version AFAIK Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* avcodec/jpeg2000dwt: also test 9/7 float DWTMichael Niedermayer2015-06-211-0/+20
| | | | Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* avcodec/jpeg2000dwt: List differences for DWT testMichael Niedermayer2015-06-211-0/+40
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
OpenPOWER on IntegriCloud