Ceres Solver

1–30 of 1390

Ceres Solver (www.ceres-solver.org) is an open source C++ library for modeling and solving large, complicated optimization problems. It can be used to solve Non-linear Least Squares problems with bounds constraints and general unconstrained optimization problems. It is a mature, feature rich, and performant library that has been used in production at Google since 2010.

This is the mailing list for discussion, bug reports etc.

0 selected

Peter Friess, … Sergiu Deitsch7

May 27

Problems with Examples

Unless you are compiling Ceres sources outside of its CMake workspace, setting the macro should not

unread,

Problems with Examples

Unless you are compiling Ceres sources outside of its CMake workspace, setting the macro should not

May 27

AD, Dmitriy Korchemkin4

May 27

Reporjection error.

Yes, if you pass the same data pointer into two different AddResidualBlock calls - those residual

unread,

Reporjection error.

Yes, if you pass the same data pointer into two different AddResidualBlock calls - those residual

May 27

Arne Jacobs, … Rasmus15

May 19

Conditional evaluation in auto-diff cost function (e.g., if-statement) possible?

Yeah actually I'm adapting a calibration solution utilizing spheres to work with boxes, because

unread,

Conditional evaluation in auto-diff cost function (e.g., if-statement) possible?

Yeah actually I'm adapting a calibration solution utilizing spheres to work with boxes, because

May 19

Robin Rofallski, … Sameer Agarwal5

May 17

nvcc error in 2.2.0 build - Visual Studio/MSVC

Ty, Alex or Sergiu know cmake well. They will review. On Wed, May 17, 2023, 1:29 AM Jason <jamak3@

unread,

nvcc error in 2.2.0 build - Visual Studio/MSVC

Ty, Alex or Sergiu know cmake well. They will review. On Wed, May 17, 2023, 1:29 AM Jason <jamak3@

May 17

Arrigo Benedetti, Dmitriy Korchemkin3

May 8

Using ceres for penalized spline regression

Good idea, thanks! On Saturday, May 6, 2023 at 12:26:19 PM UTC-7 Dmitriy Korchemkin wrote: I think

unread,

Using ceres for penalized spline regression

Good idea, thanks! On Saturday, May 6, 2023 at 12:26:19 PM UTC-7 Dmitriy Korchemkin wrote: I think

May 8

Dkhold, … Sameer Agarwal9

Apr 17

Scaling of non-linear LS shift vector for parameter weighting

Small correction to my above code: I mentioned a small but non-zero multiplier (1/1000) for

unread,

Scaling of non-linear LS shift vector for parameter weighting

Small correction to my above code: I mentioned a small but non-zero multiplier (1/1000) for

Apr 17

Estelle Jiao, Sameer Agarwal2

Apr 6

Error in evaluating the ResidualBlock -> some values are Uninitialized

The problem is that the jacobian for your cost function is 3x7 but you are only filling 18 out of

unread,

Error in evaluating the ResidualBlock -> some values are Uninitialized

The problem is that the jacobian for your cost function is 3x7 but you are only filling 18 out of

Apr 6

Dong Ren, Sergiu Deitsch2

Apr 3

Link Error with using CostFunctionToFunctor

You don't seem to be linking against Ceres which is the usual cause for such errors. There could

unread,

Link Error with using CostFunctionToFunctor

You don't seem to be linking against Ceres which is the usual cause for such errors. There could

Apr 3

Salih Marangoz, Sameer Agarwal2

Mar 27

How to interface with AutoDiffCostFunction, Nand Manifold

Salih, What you are proposing is not possible with existing tools but you can easily compose a

unread,

How to interface with AutoDiffCostFunction, Nand Manifold

Salih, What you are proposing is not possible with existing tools but you can easily compose a

Mar 27

Ghasem Abdi, … Dmitriy Korchemkin6

Mar 24

Adding Inner Constraints

Thanks Dmitriy, it makes a lot of sense. The one that you offered is the minimum constraints bundle

unread,

Adding Inner Constraints

Thanks Dmitriy, it makes a lot of sense. The one that you offered is the minimum constraints bundle

Mar 24

Joseph DeCunha, Sameer Agarwal3

Mar 23

Implementing inequality constraints by returning false from cost function

Thanks for your fast reply! I will try implementing some penalty terms in the cost function rather

unread,

Implementing inequality constraints by returning false from cost function

Thanks for your fast reply! I will try implementing some penalty terms in the cost function rather

Mar 23

James Bruce, … Sameer Agarwal7

Mar 21

Abseil logging library

Jim, I think it would make sense to wait till we move to abseil so that we are not churning on the

unread,

Abseil logging library

Jim, I think it would make sense to wait till we move to abseil so that we are not churning on the

Mar 21

Mitja P, Sameer Agarwal2

Mar 20

How to prevent scaling of entire problem to zero

Mitja, On Mon, Mar 20, 2023 at 11:34 AM Mitja P <mit...@gmail.com> wrote: I am trying to

unread,

How to prevent scaling of entire problem to zero

Mitja, On Mon, Mar 20, 2023 at 11:34 AM Mitja P <mit...@gmail.com> wrote: I am trying to

Mar 20

Johannes Brünger, Sameer Agarwal2

Mar 16

How to weight different residuals properly in ceres-solver?

yes this is reasonable. Sameer On Thu, Mar 16, 2023 at 6:57 AM Johannes Brünger <johannes.bruenger

unread,

How to weight different residuals properly in ceres-solver?

yes this is reasonable. Sameer On Thu, Mar 16, 2023 at 6:57 AM Johannes Brünger <johannes.bruenger

Mar 16

mengk...@gmail.com, Sameer Agarwal3

Mar 15

enforce different parameters to be the same result

Hi Sameer, thank you very much for your reply! Sure! That's good to know for me now, I will use a

unread,

enforce different parameters to be the same result

Hi Sameer, thank you very much for your reply! Sure! That's good to know for me now, I will use a

Mar 15

albert rasinski, Sameer Agarwal5

Mar 6

DynamicNumericDiffCostFunction with initial parameter value near 0. Step is too small.

Sorry for the delayed reply, but this seems like that the underlying library is doing something funky

unread,

DynamicNumericDiffCostFunction with initial parameter value near 0. Step is too small.

Sorry for the delayed reply, but this seems like that the underlying library is doing something funky

Mar 6

LBerger, Sameer Agarwal4

Mar 6

ceres-solver with suitesparse on windows

Laurent, Are you saying your problem is solved now? Sameer On Mon, Mar 6, 2023 at 6:10 AM LBerger

unread,

ceres-solver with suitesparse on windows

Laurent, Are you saying your problem is solved now? Sameer On Mon, Mar 6, 2023 at 6:10 AM LBerger

Mar 6

Georgios, Dmitriy Korchemkin3

Feb 21

Different values between shared and local computations

Ok, that makes sense. I will try a quaternion with a manifold instead. Thanks! On Monday, February 20

unread,

Different values between shared and local computations

Ok, that makes sense. I will try a quaternion with a manifold instead. Thanks! On Monday, February 20

Feb 21

Salih Marangoz, Sameer Agarwal5

Feb 19

Constraining Intermediate Values?

Oh, that's a good idea. Thanks! Salih On Saturday, 18 February 2023 at 01:11:41 UTC+1 sameer...@

unread,

Constraining Intermediate Values?

Oh, that's a good idea. Thanks! Salih On Saturday, 18 February 2023 at 01:11:41 UTC+1 sameer...@

Feb 19

George Chan, Sameer Agarwal2

Feb 16

OpenMP for Ceres and Eigen

This seems to be a problem with your compiler or some sort of linking mess. i would try and run the

unread,

OpenMP for Ceres and Eigen

This seems to be a problem with your compiler or some sort of linking mess. i would try and run the

Feb 16

George Chan, Sameer Agarwal2

Feb 11

WarningMessage::Allocating values array with 8 bytes

You are using miniglog which does not give you control over where the logs produced by ceres go.

unread,

WarningMessage::Allocating values array with 8 bytes

You are using miniglog which does not give you control over where the logs produced by ceres go.

Feb 11

Goksan Isil (Alumni), Sameer Agarwal3

Feb 5

Ceres Autodiff for Kalman filter tuning

I cannot access the link. On Wed, Jan 25, 2023 at 3:10 AM 'Goksan Isil (Alumni)' via Ceres

unread,

Ceres Autodiff for Kalman filter tuning

I cannot access the link. On Wed, Jan 25, 2023 at 3:10 AM 'Goksan Isil (Alumni)' via Ceres

Feb 5

Roger Labbe, Sameer Agarwal2

Jan 30

are manifold boxplus boxminus constraints numerical?

Roger, Ceres internally does not check this constraint (except for some test helpers which my check

unread,

are manifold boxplus boxminus constraints numerical?

Roger, Ceres internally does not check this constraint (except for some test helpers which my check

Jan 30

Александр Иванов, Sameer Agarwal6

Jan 17

Plan to move to 64bit indexing

Alexander, I think your idea for first making the ceres code safe wrt `-Wshorten-64-to-32` is a good

unread,

Plan to move to 64bit indexing

Alexander, I think your idea for first making the ceres code safe wrt `-Wshorten-64-to-32` is a good

Jan 17

Helen, jkfl...@gmail.com3

Jan 13

Modelling nearest distance between two 3D meshes.

Thank you for your time, and apologies, perhaps I over-simplified the problem-space! So for full

unread,

Modelling nearest distance between two 3D meshes.

Thank you for your time, and apologies, perhaps I over-simplified the problem-space! So for full

Jan 13

tingda zhuang, Ron S.2

Jan 13

Is there a quick way to modify the SetParameterization in 1.12 version to support repeated calls

I don't think there's an easy way. I believe you'd have to create a new problem and add

unread,

Is there a quick way to modify the SetParameterization in 1.12 version to support repeated calls

I don't think there's an easy way. I believe you'd have to create a new problem and add

Jan 13

Cameron Conroy, … Sameer Agarwal8

Jan 12

Ceres 2.1.0 CUDA Dense Linear Algebra Library causes preprocessor to take a VERY LONG TIME

You are solving a tiny problem (your jacobian is 8x12). Going to the gpu is not going to do anything

unread,

Ceres 2.1.0 CUDA Dense Linear Algebra Library causes preprocessor to take a VERY LONG TIME

You are solving a tiny problem (your jacobian is 8x12). Going to the gpu is not going to do anything

Jan 12

Александр Иванов, Sameer Agarwal2

Jan 9

C++ style guide

These are mostly historical artifacts from earlier versions of the code inside google. as for

unread,

C++ style guide

These are mostly historical artifacts from earlier versions of the code inside google. as for

Jan 9

Niko Haaraniemi, … Sameer Agarwal7

Jan 2

Bundle adjustment weighting with GCP variance information

Hey yzhyu, I personally weighted also the GCPs' image observation residuals with the value of 20.

unread,

Bundle adjustment weighting with GCP variance information

Hey yzhyu, I personally weighted also the GCPs' image observation residuals with the value of 20.

Jan 2

Ursela Barteczko, … Sameer Agarwal11

12/12/22

CUDA support for curve fitting

Don't switch to numeric differentiation. Make sure that you are building with optimizations

unread,

CUDA support for curve fitting

Don't switch to numeric differentiation. Make sure that you are building with optimizations

12/12/22

Search

Clear search

Close search

Google apps

Main menu