bayesian-object-tracking / dbot Goto Github PK

First off, thank you for taking the time to publish this code and the datasets!

Just a minor issue: It appears that doxygen/theme does not have a submodule mapping in .gitmodules.
Would it be possible to post a link to this submodule, and possibly include it in .gitsubmodules?

Dera Jan,

where can we find the depth based tracking dataset from your paper:
"Depth-based Object Tracking Using a Robust Gaussian Filter " ?

Thanks

the cudaMemset d_log_likelihoods error check throws the following error

Cuda error: cudaMemset d_log_likelihoods: invalid argument.

setting d_log_likelihoods requires a valid nr_poses_ value which seems to have a random value at this point because of possible uninitialized stage.

Dear Jan,

This is my setup:
gcc version 5.4.0
Eigen version 3.3.4
Ros Kinetic
CUDA Version 8.0.61

I followed the getting started README.md, but unfortunately, it doesn't compile with this setup.
I get this following error:

/usr/include/eigen3/Eigen/src/Core/CoreEvaluators.h:960:8: error: ‘Alignment’ is not a member of ‘Eigen::internal::evaluator<Eigen::VectorBlock<Eigen::Matrix<double, -1, 1>, 12> >’
/usr/include/eigen3/Eigen/src/Core/CoreEvaluators.h:984:40: error: ‘Flags’ is not a member of ‘Eigen::internal::evaluator<Eigen::VectorBlock<Eigen::Matrix<double, -1, 1>, 12> >’
/usr/include/eigen3/Eigen/src/Core/CoreEvaluators.h:960:8: error: ‘CoeffReadCost’ is not a member of ‘Eigen::internal::evaluator<Eigen::VectorBlock<Eigen::VectorBlock<Eigen::Matrix<double, -1, 1>, 12>, 6> >’
fl/include/fl/distribution/gaussian.hpp:464:37: error: no match for ‘operator-’ (operand types are ‘fl::Real {aka double}’ and ‘Eigen::internal::enable_if<true, const Eigen::CwiseBinaryOp<Eigen::internal::scalar_product_op<double, double>, const Eigen::CwiseNullaryOp<Eigen::internal::scalar_constant_op<double>, const Eigen::Matrix<double, 1, 1, 0, 1, 1> >, const Eigen::CwiseBinaryOp<Eigen::internal::scalar_difference_op<double, double>, const Eigen::Matrix<double, 1, 1, 0, 1, 1>, const Eigen::Matrix<double, 1, 1, 0, 1, 1> > > >::type {aka const Eigen::CwiseBinaryOp<Eigen::internal::scalar_product_op<double, double>, const Eigen::CwiseNullaryOp<Eigen::internal::scalar_constant_op<double>, const Eigen::Matrix<double, 1, 1, 0, 1, 1> >, const Eigen::CwiseBinaryOp<Eigen::internal::scalar_difference_op<double, double>, const Eigen::Matrix<double, 1, 1, 0, 1, 1>, const Eigen::Matrix<double, 1, 1, 0, 1, 1> > >}’) return log_normalizer() - 0.5

Do you have any idea or solution for this kind of error?

Regards

There is a an issue with allocating and copying occlusion probability vector