Fix: error in the dimension used for computeMinMax#17607
Fix: error in the dimension used for computeMinMax#17607pemmanuelviel wants to merge 1 commit intoopencv:masterfrom
Conversation
Instead of using the current dimension for which we just got a big span, we were computing Min and Max for the previous dimension stored in cutfeat (and using 0 instead of the dimension indice for the very first dimension with "span > (1-eps)max_span")
|
@pemmanuelviel Thanks for the patch. Could you provide small test that proofs your fix? |
|
@pemmanuelviel Also it looks like the same issue is present in 3.4 branch too. This patch should go into 3.4 branch first. We will merge changes from 3.4 into master regularly (weekly/bi-weekly). So, please:
Note: no need to re-open PR, apply changes "inplace". |
|
@pemmanuelviel, thank you, looks good to me! 👍 |
|
Hi @vpisarev , |
|
@asmorkalov To ensure each commit is fine on 3.4 I've created new branches (replacing pev-xxx by pev--xxx) and cherry-picked each commit. These new branches are currently in pull-request. |
|
BTW, how can I associate labels to my PR? Can't find where in the gui... |
|
@pemmanuelviel Only maintainers add labels to PRs. |
Instead of using the current dimension for which we just got a big span,
we were computing Min and Max for the previous dimension stored in cutfeat
(and using 0 instead of the dimension indice for the very first dimension
with "span > (1-eps)max_span")
Pull Request Readiness Checklist
See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request
Patch to opencv_extra has the same branch name.