Mitja P
Mitja P
Sum should be recalculated after inverting scores as sum is not the same anymore. Application crashes when adding new scores until it reaches _level_ (which is between 0 and sum,...
I might have found a potential issue with unaligned Eigen objects and `std::vector`. I think the issue is with `Half_plane` and `Half_planes` definitions https://github.com/openMVG/openMVG/blob/5e98d504bb76ba2d1d07ae80ac2acb10b3d6f97d/src/openMVG/geometry/half_space_intersection.hpp#L27 There are some `std::vector` specializations https://github.com/openMVG/openMVG/blob/74deb33d12bf275a3b3a9afc833f4760be90f031/src/openMVG/numeric/eigen_alias_definition.hpp#L142...
I'm proposing some changes to CMake that would also install config file which makes it easier for downstream libraries to use avro as dependency. Make sure you have checked _all_...
When loading images you apply gamma expension (correction) with following code: https://github.com/Erkaman/poisson_blend/blob/94e6a4ad18876848c631f64184d37e69584477c7/src/main.cpp#L69 But this is actually gamma compression. The correct way for gamma expansion would be ``` pow(buf[i + 0]...
Shouldn't this check bi inside of `ModeFunctions::refine` function? https://github.com/alicevision/popsift/blob/fafcad973bd7b1740acd4c276922fe039516a787/src/popsift/s_extrema.cu#L447 There is already a comment in place suggesting this: https://github.com/alicevision/popsift/blob/fafcad973bd7b1740acd4c276922fe039516a787/src/popsift/s_extrema.cu#L160 Hope I'm not nitpicking here :) Just trying to make code...
Datastores's responsibility should not be handling of http requests but working with readable streams and File class. This PR is just a first step towards this goal, it is probably...
The following example copies detection data in a pointer type (`apriltag_detection_t *`) instead to value type (`apriltag_detection_t`). This causes the `zarray_get` to write to memory on stack outside its bounds...
I have implemented `PointToPointFunctor` and `PointToPlaneFunctor` (default) and introducted new optional dependency on Ceres. If ceres is not found library uses previous linear method. Is this OK?
I've noticed that in some places you used 1-based indices. I guess all of this should be 0-based? https://github.com/pglira/simpleICP/blob/1478851961c01eab677bd325f93dff9b85d7a069/c%2B%2B/src/simpleicp.cpp#L164 https://github.com/pglira/simpleICP/blob/1478851961c01eab677bd325f93dff9b85d7a069/c%2B%2B/src/simpleicp.cpp#L181 https://github.com/pglira/simpleICP/blob/1478851961c01eab677bd325f93dff9b85d7a069/c%2B%2B/src/pointcloud.cpp#L82
When building transformation matrix from 'alpha1', 'alpha2', 'alpha3', rotation part is not correct. 'alpha1', 'alpha2', 'alpha3' are Euler angles and can not be just inserted in matrix like this: https://github.com/pglira/simpleICP/blob/fd1a926bc0b605c658dad08acd842e433f5eacf7/c%2B%2B/src/corrpts.cpp#L133...