Pith. sign in

REVIEW 1 major objections 4 minor 44 references

Well-posedness of an optical flow based optimal control formulation for image registration

T0 review · 1 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper shows that smoothed Orlicz-space relaxations make Lipschitz-regularized optical-flow image registration well-posed.

desk verdict Solid, genuinely useful theory paper: new transport uniqueness in exponential Orlicz spaces plus a clean relaxation of W^{1,\infty} regularization; deserves a serious referee. read the letter →

arxiv 2507.10188 v4 pith:3KNFANAB submitted 2025-07-14 math.OC

Johannes Haubner , Christian Clason This is my paper · ORCID
classification math.OC MSC 49J2035L0446E3065N30
keywords imageregistrationopticalflowoptimalcontrollineartransportequationOrliczspacesW^{1∞}regularizationwell-posednessfiniteelementsemidiscretization
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper establishes the well-posedness of an optimal control formulation of image registration in which the image is transported by a velocity field and the regularization penalizes the Lipschitz ($W^{1,\infty}$) norm of that field. Since $W^{1,\infty}$ is non-reflexive, standard compactness arguments do not apply, and the paper replaces the Lipschitz norm by a smoothed functional built from the exponential and the Orlicz space $L^{\exp}$. The core results are that the smoothed problems have global minimizers, that their finite-element semidiscretizations are consistent, and that as the smoothing parameter tends to zero the discrete minimizers converge to a global minimizer of the original $W^{1,\infty}$-regularized problem. The proof requires a new uniqueness theorem for the linear transport equation when the divergence of the velocity lies only in $L^p((0,T),L^{\exp}(\Omega))$; earlier results needed bounded divergence. A sympathetic reader would care because this supplies a missing theoretical foundation for a widely used registration model that previously lived between too-regular Hilbert settings and settings without guarantees.

What carries the argument

The load-bearing object is the smoothed essential-supremum functional $\chi_\gamma(u)=\gamma\ln\bigl(\tfrac{1}{|\Omega|}\int_\Omega \exp(\gamma^{-1}u)\bigr)+\gamma\ln\bigl(\tfrac{1}{|\Omega|}\int_\Omega \exp(-\gamma^{-1}u)\bigr)$, which tends to $\operatorname{ess\,sup}|u|$ as $\gamma\to0$ and is naturally posed on the Orlicz space $L^{\exp}(\Omega)$. Its dual representation as a supremum over $L\log L$ densities with a relative-entropy barrier supplies convexity, monotone dependence on $\gamma$, and weak-star lower semicontinuity. The transport equation is handled through the renormalization property for $BV/L^\infty$ velocities, and a Gr\"onwall-type inequality applied on short time intervals to the function $\exp(\gamma^{-1}\phi^2+1)$ proves uniqueness with only $L^p_t L^{\exp}$ divergence regularity. Together these pieces verify the fifteen abstract properties of the paper's conceptual framework, which is what transfers compactness and limit arguments from the abstract setting to the image registration application.

What would settle it

Look for a velocity field $b$ in $L^1((0,T),BV_0)^d \cap L^\infty((0,T)\times\Omega)^d$ with $\operatorname{div} b\in L^p((0,T),L^{\exp}(\Omega))$ but not in $L^p((0,T),L^\infty(\Omega))$ for which the transport equation admits two weak solutions in $C([0,T],L^\infty\text{-}w^{\ast})$; exhibiting one would refute Theorem 5.9 and collapse the paper's well-posedness chain. A computational check would be to solve the semidiscretized relaxed problem for a pair of images with $\sigma=0$ and compare the $\gamma\to0$ limit with the exact $W^{1,\infty}$ minimizer, since failure there would show the $H^{1+\sigma}$ bottleneck is substantive.

Watch

Extended reading notes

Core claim

The paper's central claim is that the relaxed image registration problems are well-posed and approximate the original Lipschitz-regularized problem exactly. Concretely, for $\sigma\in(0,\tfrac12)$, the paper takes admissible velocities in $H^{1+\sigma}(\Omega)^d \cap W^{1,\infty}_0(\Omega)^d$ and shows that for every $\gamma>0$ the problem $\min_v J(\phi(\cdot,T),\phi_{\mathrm{tar}})+\beta\Psi_\gamma(v)+\tfrac{\alpha}{2}\|v\|_{H^{1+\sigma}}^2$ subject to $\partial_t\phi+v\cdot\nabla\phi=0$ has a solution; Corollary 6.4 then provides a sequence $\gamma_k\to 0$ such that minimizers of the $P1$ finite-element semidiscretizations converge, in the relevant weak and weak-star topologies, to a global minimizer of the original $W^{1,\infty}$-regularized problem. The transport-equation engine is Theorem 5.9, which guarantees a unique weak solution in $C([0,T],L^{\infty}\text{-}w^{\ast})$ whenever $b\in L^1((0,T),BV_0)^d\cap L^\infty((0,T)\times\Omega)^d$ and $\operatorname{div} b\in L^p((0,T),L^{\exp}(\Omega))$ with $p\in(1,\infty]$, relaxing the earlier requirement that the divergence be bounded.

Load-bearing premise

The load-bearing premise is that admissible stationary velocities lie in $H^{1+\sigma}(\Omega)^d$ with $\sigma>0$, which forces them into $L^\infty$; at the natural $\sigma=0$ endpoint in two dimensions, $H^1$ does not embed into $L^\infty$, and the transport uniqueness and relaxation-limit results as stated do not cover the relaxed problem.

Editorial extensions

If this is right

  • For every positive smoothing parameter $\gamma$, the relaxed registration problem has a global minimizer in $H^{1+\sigma}(\Omega)^d \cap W^{1,\exp}_0(\Omega)^d$, so computations on finite elements start from a well-posed problem.
  • Along a deliberately chosen sequence $\gamma_k\to0$, minimizers of the semidiscretized relaxed problems have subsequences that converge to a global minimizer of the $W^{1,\infty}$-regularized problem, so the smoothing does not change the model in the limit.
  • The transport equation with velocity in $L^1_t BV_0 \cap L^\infty$ and divergence only in $L^p_t L^{\exp}$ has a unique weak solution, extending earlier uniqueness theory that required bounded divergence.
  • The smoothed norm is Fr\'echet differentiable on $L^\infty$ and, on finite-dimensional spaces, gives a differentiable surrogate for the nonsmooth $W^{1,\infty}$ norm, removing a differentiability obstruction to numerical optimization.
  • Because the framework is abstract, the same pattern of relaxation, discretization, and limit applies to any other control problem satisfying the same hypotheses, not only optical-flow registration.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • I would expect the same entropic smoothing to transfer to other $W^{1,\infty}$-regularized problems, such as shape optimization with bi-Lipschitz transformations, where the same non-reflexive obstruction appears.
  • The reliance on $\sigma>0$ suggests that numerical experiments at $\sigma=0$ might reveal qualitatively different behavior; testing whether the relaxation limits still converge there would map the boundary of the theory.
  • Because $\chi_\gamma$ is the entropic risk measure, the relaxation could be interpreted as optimizing under an exponential utility, which may offer a stochastic or robust-control reading of the registration model.
  • A natural next step would be to check whether the stability result survives when the strong convergence assumptions on $b_n$ and $\operatorname{div} b_n$ in Proposition 5.11 are weakened to the bounded-and-weak conditions used in the bounded-divergence case, which Remark 5.12 leaves open.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

1 major / 4 minor

Summary. The paper studies the optimal-control formulation of image registration, where the state equation is the linear transport equation and the control is regularized by a W^{1,\infty} norm. Because this norm is non-smooth and the natural space is non-reflexive, the authors replace it by an entropic relaxation \Psi_\gamma defined through Orlicz spaces L^exp, prove a uniqueness theorem for the transport equation when div b \in L^p((0,T), L^exp), establish a stability result, and embed the whole construction in an abstract framework (p1)–(p15) that yields existence of minimizers for the relaxed problem, convergence of semidiscrete FE relaxations, and Gamma-type convergence as \gamma\to 0. The main application is Corollary 6.4, which asserts that a sequence of discrete relaxed minimizers converges to a global minimizer of the original W^{1,\infty}-regularized problem.

Significance. The main contribution is a missing well-posedness foundation for Lipschitz-regularized optical-flow registration. Theorem 5.9 is a genuine extension of earlier uniqueness results: it replaces div b \in L^\infty in time by div b \in L^p((0,T), L^exp(\Omega)), and its proof via renormalization, Orlicz duality, and Gr\u00f6nwall's inequality is internally consistent. The abstract framework isolates the hypotheses cleanly, and the paper is careful to state which properties are proved where. I also note that the stationary controls used in Section 6 automatically satisfy the L^\infty hypothesis of Theorem 5.9: W^{1,exp}_0 embeds into W^{1,p}_0 for every p<\infty and hence into L^\infty; the role of H^{1+\sigma} is to make X compactly embedded into W=W^{1,1}, not to ensure boundedness of controls. The only substantive defect I found is a repairable gap in the stability proof in Appendix B; there is no circularity or hidden data-fitting in the argument.

major comments (1)
  1. [Appendix B, Eq. (b.4)] The proof of Proposition 5.11 defines C_2 := sup_{n\in\mathbb{N}}(\|b_n(t,\cdot)\|_{L^1(\Omega)^d} + \|\mathrm{div}(b_n(t,\cdot))\|_{L^1(\Omega)}) and states that C_2 is bounded "due to properties (ii) and (iii)". Those properties only give strong convergence in L^1((0,T),L^1(\Omega)) or L^1((0,T)\times\Omega), which does not imply an a.e.-in-time uniform bound of the form used in (b.4). The pointwise estimate is therefore not justified. This is not merely cosmetic: the equicontinuity step used in the Arzel\u00e0–Ascoli argument depends on this bound, and Proposition 5.11 is the basis for property (p7) in Lemma 6.1 and hence for Corollaries 6.2 and 6.4. The repair is standard: use that t \mapsto \|b_n(t,\cdot)\|_{L^1} + \|\mathrm{div}(b_n(t,\cdot))\|_{L^1} is uniformly integrable in t, choose \delta so that \int_{t_1}^{t_2}(\|b_n(t,\cdot)\|_{L^1} + \|\mathrm{div}(b_n(t,\cdot))\|_{L^1})\,dt \le \varepsilon/(C_1\|\varphi_k\|_{H^1}) for all n whenever |t_2-t_1|<\delta, and insert this bound in (b.4). With this replacement the proof of Proposition 5.11 goes through.
minor comments (4)
  1. [Section 5.2, Eq. (5.12)] In Eq. (5.12) the left-hand side should be \int_\Omega \beta(\phi(\tau))\,dx with \tau = \min(r+\bar t,T), and the interval in the subsequent Gr\u00f6nwall estimate should be [r, \min(r+\bar t,T)] instead of [0,\bar t]. The iterative argument is clear from context, but the display as written is inconsistent.
  2. [Section 6.3, Lemma 6.3] The extension map E is required to satisfy both E(u)|_{\partial\Omega} = u|_{\partial\Omega} and E(u)\in C_c^\infty(\Omega); these requirements are incompatible unless u|_{\partial\Omega}=0. It should presumably be E(u)\in C^\infty(\overline{\Omega}) or a similar smoothness condition; once that typo is fixed, the construction of \check w works as intended.
  3. [Cross-references] The running text refers to "Sectiona" (page 17), "Sectionb" (page 21), and "appendixc" (page 22); these should be replaced by proper cross-references to Appendices A, B, and C, respectively.
  4. [Appendix B] In the proof of Proposition 5.11, the sentence "for almost every \varphi\in L^2(\Omega)" should read "for every \varphi\in L^2(\Omega)", and the display in (b.3) contains a typo in the middle term ("\varphi_k - \varphi_2").

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the central theorems are proven from stated assumptions and external results by non-overlapping authors; the relaxation limit is derived, not assumed.

full rationale

The paper is a self-contained mathematical analysis. The main theorems (Theorem 5.9, Corollary 5.10, Proposition 5.11, Corollary 6.4) are proved from explicitly stated assumptions using standard tools such as Grönwall's inequality, Orlicz-space duality, and the direct method. The relaxation functional Ψ_γ is defined explicitly in (4.19), and its convergence to the W^{1,∞} norm is proven in Lemma 4.7 from elementary asymptotic and convexity estimates; no parameter is fitted to data and no quantity is renamed a prediction. The abstract framework of Section 2 assumes properties (p1)–(p15), but each property is later verified for the image-registration setting in Lemmas 6.1 and 6.3; the gamma-convergence arguments in Lemmas 2.6–2.8 and Theorem 2.9 are direct proofs rather than citations. The external results cited for the transport equation, mainly [28] and [29], are by Jarde and Jarde–Ulbrich and have no author overlap with the present paper; they are used as independent mathematical facts. The self-citations [14] and [46] are contextual — a known regularization technique and a formulation reference — and are not load-bearing for the existence, uniqueness, or convergence claims. The only noted technical blemish, a uniform-integrability constant in Appendix B, is a correctness detail and not a circular reduction.

Assumptions & free parameters 0 free parameters · 6 assumptions · 0 invented entities

The paper contributes a mathematical proof framework and introduces no empirically fitted parameters and no new physical or mathematical entities. The constants α, β, γ, and σ are model parameters whose specific values are irrelevant to the universal well-posedness statement. All background results are from the standard functional analysis, Orlicz space, and transport theory literature, used as benchmarks rather than assumed conclusions.

assumptions (6)
  • standard math Renormalization property for BV ∩ L∞ velocity fields (Prop 5.7, from [28, Thm 3.1.26])
    Used in the proof of Theorem 5.9 to convert weak solutions into renormalized solutions, a linchpin of the uniqueness argument.
  • standard math Existence theorem for transport equations with BV and L1 divergence (Prop 5.5, from [29])
    Provides existence of weak solutions in C([0,T], L^∞(Ω)-w*), on which uniqueness and the reduced objective rely.
  • standard math L^exp is the dual of separable L^{log L}, and W^{1,exp}_0 has a separable predual (Remark 3.8 and Appendix C)
    Load-bearing for weak-* compactness arguments (properties p2, p5, p9) in the direct method.
  • domain assumption Sobolev embedding H^{1+σ}(Ω) ↪ L∞(Ω) for σ > 0 and d ∈ {2,3}
    Ensures the stationary velocity fields in the image registration application are in L∞((0,T)×Ω), a requirement of the transport uniqueness theorem; fails at σ = 0 for d = 2.
  • domain assumption Smooth boundary of Ω in Lemma 6.3 for the finite-element approximation property (p11)
    Needed for the extension/reflection operator and interpolation estimates; the rest of the paper only requires a bounded Lipschitz domain.
  • standard math Donsker-Varadhan variational formula from [11] (Lemma 4.1)
    Establishes the dual representation of the smoothed essential supremum, used for weak-* lower semicontinuity and the γ→0 limit.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Well-posedness of an optical flow based optimal control formulation for image registration." pith.science (2026). https://pith.science/paper/3KNFANAB

@misc{pith2026250710188,
  author       = {Pith},
  title        = {Pith review of: Well-posedness of an optical flow based optimal control formulation for image registration},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3KNFANAB}},
  note         = {Machine review of arXiv:2507.10188}
}
read the original abstract

We consider image registration as an optimal control problem using an optical flow formulation, i.e., we discuss an optimization problem that is governed by a linear hyperbolic transport equation. Requiring Lipschitz continuity of the vector fields that parametrize the transformation leads to an optimization problem in a non-reflexive Banach space. We introduce relaxations of the optimization problem involving smoothed maximum and minimum functions and appropriate Orlicz spaces. To derive well-posedness results for the relaxed optimization problem, we revisit and establish new existence and uniqueness results for the linear hyperbolic transport equations. We further discuss limit considerations with respect to the relaxation parameter and discretizations.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

44 extracted references · 32 canonical work pages

  1. [1]

    Acar and C

    R. Acar and C. R. Vogel, Analysis of bounded variation penalty methods for ill-posed prob- lems,Inverse Problems10 (1994), 1217–1229,doi:10.1088/0266-5611/10/6/003,https://doi.org/10.1088/ 0266-5611/10/6/003

    work page doi:10.1088/0266-5611/10/6/003 1994

  2. [2]

    Ambrosio, Transport equation and Cauchy problem for 𝐵𝑉 vector fields,Invent

    L. Ambrosio, Transport equation and Cauchy problem for 𝐵𝑉 vector fields,Invent. Math.158 (2004), 227–260,doi:10.1007/s00222-004-0367-2,https://doi.org/10.1007/s00222-004-0367-2

    doi:10.1007/s00222-004-0367-2 2004

  3. [3]

    Ambrosio and A

    L. Ambrosio and A. Figalli, On flows associated to Sobolev vector fields in Wiener spaces: an approach à la DiPerna-Lions,J. Funct. Anal.256 (2009), 179–214,doi:10.1016/j.jfa.2008.05.007, https://doi.org/10.1016/j.jfa.2008.05.007

    doi:10.1016/j.jfa.2008.05.007 2009

  4. [4]

    Ben Belgacem and S

    F. Ben Belgacem and S. C. Brenner, Some nonstandard finite element estimates with applications to 3D Poisson and Signorini problems,Electron. Trans. Numer. Anal.12 (2001), 134–148

    work page 2001

  5. [5]

    Bennett and R

    C. Bennett and R. Sharpley,Interpolation of operators, volume 129 of Pure and Applied Mathematics, Academic Press, Inc., Boston, MA, 1988

    work page 1988

  6. [7]

    Borzi, K

    A. Borzi, K. Ito, and K. Kunisch, Optimal control formulation for determining optical flow, SIAM J. Sci. Comput.24 (2002), 818–847,doi:10.1137/s1064827501386481,https://doi.org/10.1137/ S1064827501386481

    work page doi:10.1137/s1064827501386481 2002

  7. [8]

    Bouchut, F

    F. Bouchut, F. James, and S. Mancini, Uniqueness and weak stability for multi-dimensional trans- port equations with one-sided Lipschitz coefficient,Ann. Sc. Norm. Super. Pisa Cl. Sci. (5)4 (2005), 1–25

    work page 2005

  8. [9]

    Brandenburg, F

    C. Brandenburg, F. Lindemann, M. Ulbrich, and S. Ulbrich, A continuous adjoint approach to shape optimization for Navier Stokes flow, inOptimal control of coupled systems of partial differ- ential equations, volume 158 of Internat. Ser. Numer. Math., Birkhäuser Verlag, Basel, 2009, 35–56, doi:10.1007/978-3-7643-8923-9_2,https://doi.org/10.1007/978-3-7643-8923-9_2

    work page doi:10.1007/978-3-7643-8923-9_2 2009

Show all 44 references
  1. [10]

    Bredies and M

    K. Bredies and M. Holler, A pointwise characterization of the subdifferential of the total variation functional (2016),arXiv:1609.08918

    2016 arXiv

  2. [11]

    Budhiraja and P

    A. Budhiraja and P. Dupuis,Analysis and approximation of rare events, volume 94 of Probabil- ity Theory and Stochastic Modelling, Springer, New York, 2019,doi:10.1007/978-1-4939-9579-0. Representations and weak convergence methods

    2019 doi

  3. [12]

    V. V. Buldygin and Y. V. Kozachenko,Metric characterization of random variables and random processes, volume 188 of Translations of Mathematical Monographs, American Mathematical Society, Providence, RI, 2000,doi:10.1090/mmono/188. Translated from the 1998 Russian original by ...

    doi

  4. [13]

    Chen and D

    K. Chen and D. A. Lorenz, Image sequence interpolation using optimal control,J. Math. Imaging Vision41 (2011), 222–238,doi:10.1007/s10851-011-0274-2,https://doi.org/10.1007/s10851-011-0274-2

    2011 doi

  5. [14]

    Clason, 𝐿∞ fitting for inverse problems with uniform noise,Inverse Problems28 (2012), 104007, 18,doi:10.1088/0266-5611/28/10/104007,https://doi.org/10.1088/0266-5611/28/10/104007

    C. Clason, 𝐿∞ fitting for inverse problems with uniform noise,Inverse Problems28 (2012), 104007, 18,doi:10.1088/0266-5611/28/10/104007,https://doi.org/10.1088/0266-5611/28/10/104007

    2012 doi

  6. [15]

    Clason, K

    C. Clason, K. Ito, and K. Kunisch, A minimum effort optimal control problem for elliptic PDEs, ESAIM Math. Model. Numer. Anal.46 (2012), 911–927,doi:10.1051/m2an/2011074,https://doi.org/10. 1051/m2an/2011074

    2012

  7. [16]

    Clason, D

    C. Clason, D. A. Lorenz, H. Mahler, and B. Wirth, Entropic regularization of continuous optimal transport problems,J. Math. Anal. Appl.494 (2021), Paper No. 124432, 22,doi:10.1016/j.jmaa.2020. 124432,https://doi.org/10.1016/j.jmaa.2020.124432

    2021

  8. [17]

    E. D. Conway, Generalized solutions of linear differential equations with discontinuous coefficients and the uniqueness question for multidimensional quasilinear conservation laws,J. Math. Anal. Appl.18 (1967), 238–251,doi:10.1016/0022-247x(67)90054-6,https://doi.org/10.1016/0...

    1967 doi

  9. [18]

    Crippa,The Flow Associated to Weakly Differentiable Vector Fields, PhD thesis, University of Zurich, 2008,doi:10.5167/uzh-163873

    G. Crippa,The Flow Associated to Weakly Differentiable Vector Fields, PhD thesis, University of Zurich, 2008,doi:10.5167/uzh-163873

    2008 doi

  10. [19]

    Crippa, C

    G. Crippa, C. Donadello, and L. V. Spinolo, Initial-boundary value problems for continuity equa- tions with𝐵𝑉 coefficients,J. Math. Pures Appl. (9)102 (2014), 79–98,doi:10.1016/j.matpur.2013.11.002, https://doi.org/10.1016/j.matpur.2013.11.002

    2014 doi

  11. [21]

    De Lellis, Ordinary differential equations with rough coefficients and the renormalization theorem of Ambrosio [after Ambrosio, DiPerna, Lions], number 317, 2008, Exp

    C. De Lellis, Ordinary differential equations with rough coefficients and the renormalization theorem of Ambrosio [after Ambrosio, DiPerna, Lions], number 317, 2008, Exp. No. 972, viii, 175–203. Séminaire Bourbaki. Vol. 2006/2007

    2008

  12. [22]

    Deckelnick, P

    K. Deckelnick, P. J. Herbert, and M. Hinze, A novel𝑊 1,∞ approach to shape optimisation with Lipschitz domains,ESAIM Control Optim. Calc. Var.28 (2022), Paper No. 2, 29,doi:10.1051/cocv/ 2021108,https://doi.org/10.1051/cocv/2021108

    2022

  13. [23]

    Diehl, J

    E. Diehl, J. Haubner, M. Ulbrich, and S. Ulbrich, Differentiability results and sensitivity calculation for optimal control of incompressible two-phase Navier-Stokes equations with surface tension, Comput. Optim. Appl.87 (2024), 711–751,doi:10.1007/s10589-022-00415-6,https://d...

    2024 doi

  14. [24]

    R. J. DiPerna and P. L. Lions, Ordinary differential equations, transport theory and Sobolev spaces, Invent. Math.98 (1989), 511–547,doi:10.1007/bf01393835,https://doi.org/10.1007/BF01393835

    1989 doi

  15. [25]

    T. K. Donaldson and N. S. Trudinger, Orlicz-Sobolev spaces and imbedding theorems,J. Functional Analysis8 (1971), 52–75,doi:10.1016/0022-1236(71)90018-8,https://doi.org/10.1016/0022-1236(71) 90018-8

    1971 doi

  16. [26]

    Dupuis and R

    P. Dupuis and R. S. Ellis,A weak convergence approach to the theory of large deviations, Wiley Series in Probability and Statistics: Probability and Statistics, John Wiley & Sons, Inc., New York, 1997,doi:10.1002/9781118165904. A Wiley-Interscience Publication

    1997 doi

  17. [27]

    Gwiazda, S

    P. Gwiazda, S. C. Hille, K. Ł yczek, and A. Świerczewska-Gwiazda, Differentiability in pertur- bation parameter of measure solutions to perturbed transport equation,Kinet. Relat. Models12 (2019), 1093–1108,doi:10.3934/krm.2019041,https://doi.org/10.3934/krm.2019041

    2019 doi

  18. [28]

    Jarde,Analysis of Optimal Control Problems for the Optical Flow Equation under Mild Regularity Assumptions, PhD thesis, Technische Universität München, 2018

    P. Jarde,Analysis of Optimal Control Problems for the Optical Flow Equation under Mild Regularity Assumptions, PhD thesis, Technische Universität München, 2018

    2018

  19. [29]

    Jarde and M

    P. Jarde and M. Ulbrich, Existence of minimizers for optical flow based optimal control problems under mild regularity assumptions,Control Cybernet.48 (2019), 261–306

    2019

  20. [30]

    Kosmol and D

    P. Kosmol and D. Müller-Wichards,Optimization in function spaces, volume 13 of De Gruyter Series in Nonlinear Analysis and Applications, Walter de Gruyter & Co., Berlin, 2011. With stability considerations in Orlicz spaces

    2011

  21. [31]

    Kruse and M

    F. Kruse and M. Ulbrich, A self-concordant interior point approach for optimal control with state constraints,SIAM J. Optim.25 (2015), 770–806,doi:10.1137/130936671,https://doi.org/10.1137/ 130936671

    2015 doi

  22. [32]

    Leoni,A first course in Sobolev spaces, volume 181 of Graduate Studies in Mathematics, American Mathematical Society, Providence, RI, second edition, 2017,doi:10.1090/gsm/181

    G. Leoni,A first course in Sobolev spaces, volume 181 of Graduate Studies in Mathematics, American Mathematical Society, Providence, RI, second edition, 2017,doi:10.1090/gsm/181

    2017 doi

  23. [33]

    J. L. Lions and E. Magenes,Non-homogeneous boundary value problems and applications. Vol. I, volume Band 181 of Die Grundlehren der mathematischen Wissenschaften, Springer-Verlag, New York-Heidelberg, 1972. Translated from the French by P. Kenneth

    1972

  24. [34]

    P. L. Lions and B. Seeger, Transport equations and flows with one-sided Lipschitz velocity fields, Arch. Ration. Mech. Anal.248 (2024), Paper No. 86, 61,doi:10.1007/s00205-024-02029-0,https: //doi.org/10.1007/s00205-024-02029-0. Haubner, Clason Well-posedness of an optical flo...

    2024 arXiv

  25. [35]

    D. S. Mitrinović, J. E. Pečarić, and A. M. Fink,Inequalities involving functions and their integrals and derivatives, volume 53 of Mathematics and its Applications (East European Series), Kluwer Academic Publishers Group, Dordrecht, 1991,doi:10.1007/978-94-011-3562-7

    1991 doi

  26. [36]

    P. B. Mucha, Transport equation: extension of classical results for div𝑏∈BMO ,J. Differential Equations249 (2010), 1871–1883,doi:10.1016/j.jde.2010.07.015,https://doi.org/10.1016/j.jde.2010.07. 015

    2010 doi

  27. [37]

    L. Pick, A. Kufner, O. John, and S. Fučík,Function spaces. Vol. 1, volume 14 of De Gruyter Series in Nonlinear Analysis and Applications, Walter de Gruyter & Co., Berlin, extended edition, 2013

    2013

  28. [38]

    Ring, Structural properties of solutions to total variation regularization problems,M2AN Math

    W. Ring, Structural properties of solutions to total variation regularization problems,M2AN Math. Model. Numer. Anal.34 (2000), 799–810,doi:10.1051/m2an:2000104,https://doi.org/10.1051/m2an: 2000104

    2000 doi

  29. [39]

    Rudin,Functional analysis, International Series in Pure and Applied Mathematics, McGraw-Hill, Inc., New York, second edition, 1991

    W. Rudin,Functional analysis, International Series in Pure and Applied Mathematics, McGraw-Hill, Inc., New York, second edition, 1991

    1991

  30. [40]

    Shapiro, D

    A. Shapiro, D. Dentcheva, and A. Ruszczyński,Lectures on stochastic programming—modeling and theory, volume 28 of MOS-SIAM Series on Optimization, Society for Industrial and Applied Mathematics (SIAM), Philadelphia, PA; Mathematical Optimization Society, Philadelphia, PA, thir...

    2021 doi

  31. [41]

    Simon,Convexity, volume 187 of Cambridge Tracts in Mathematics, Cambridge University Press, Cambridge, 2011,doi:10.1017/cbo9780511910135

    B. Simon,Convexity, volume 187 of Cambridge Tracts in Mathematics, Cambridge University Press, Cambridge, 2011,doi:10.1017/cbo9780511910135. An analytic viewpoint

    2011 doi

  32. [42]

    Subko, A remark on the transport equation with𝑏∈BV and div𝑥𝑏∈BMO ,Colloq

    P. Subko, A remark on the transport equation with𝑏∈BV and div𝑥𝑏∈BMO ,Colloq. Math.135 (2014), 113–125,doi:10.4064/cm135-1-9,https://doi.org/10.4064/cm135-1-9

    2014 doi

  33. [43]

    J. Tao, D. Yang, and W. Yuan, A Survey on Function Spaces of John–Nirenberg Type,Mathematics 9 (2021),doi:10.3390/math9182264,https://www.mdpi.com/2227-7390/9/18/2264

    2021 doi

  34. [44]

    Tröltzsch,Optimal control of partial differential equations, volume 112 of Graduate Studies in Mathematics, American Mathematical Society, Providence, RI, 2010,doi:10.1090/gsm/112

    F. Tröltzsch,Optimal control of partial differential equations, volume 112 of Graduate Studies in Mathematics, American Mathematical Society, Providence, RI, 2010,doi:10.1090/gsm/112. Theory, methods and applications, Translated from the 2005 German original by Jürgen Sprekels

    2010 doi

  35. [45]

    Turett, Fenchel-Orlicz spaces,Dissertationes Math

    B. Turett, Fenchel-Orlicz spaces,Dissertationes Math. (Rozprawy Mat.)181 (1980), 55

    1980

  36. [46]

    B. Zapf, J. Haubner, L. Baumgärtner, and S. Schmidt, Medical image registration using optimal control of a linear hyperbolic transport equation with a DG discretization,SIAM J. Sci. Comput. 47 (2025), B253–B279,doi:10.1137/23m1570727,https://doi.org/10.1137/23M1570727. Haubner...

    2025 doi

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.