Search

This research paper proposes a Latent Diffusion Model for 3D (LDM3D) that generates both image and depth map data from a given text prompt, allowing users to generate RGBD images from text prompts. The LDM3D model is fine-tuned on a dataset of tuples containing an RGB image, depth map and caption, and validated through extensive experiments. We also develop an application called DepthFusion, which uses the generated RGB images and depth maps to create immersive and interactive 360-degree-view experiences using TouchDesigner. This technology has the potential to transform a wide range of industries, from entertainment and gaming to architecture and design. Overall, this paper presents a significant contribution to the field of generative AI and computer vision, and showcases the potential of LDM3D and DepthFusion to revolutionize content creation and digital experiences. A short video summarizing the approach can be found at https://t.ly/tdi2.

Control design for robotic systems is complex and often requires solving an optimization to follow a trajectory accurately. Online optimization approaches like Model Predictive Control (MPC) have been shown to achieve great tracking performance, but require high computing power. Conversely, learning-based offline optimization approaches, such as Reinforcement Learning (RL), allow fast and efficient execution on the robot but hardly match the accuracy of MPC in trajectory tracking tasks. In systems with limited compute, such as aerial vehicles, an accurate controller that is efficient at execution time is imperative. We propose an Analytic Policy Gradient (APG) method to tackle this problem. APG exploits the availability of differentiable simulators by training a controller offline with gradient descent on the tracking error. We address training instabilities that frequently occur with APG through curriculum learning and experiment on a widely used controls benchmark, the CartPole, and two common aerial robots, a quadrotor and a fixed-wing drone. Our proposed method outperforms both model-based and model-free RL methods in terms of tracking error. Concurrently, it achieves similar performance to MPC while requiring more than an order of magnitude less computation time. Our work provides insights into the potential of APG as a promising control method for robotics. To facilitate the exploration of APG, we open-source our code and make it available atgithub.com/lis-epfl/apg_trajectory_tracking.

Accurate manipulations of an open quantum system require a deep knowledge of its controllability properties and the information content of the implemented control fields. By using tools of information and quantum optimal control theory, we provide analytical bounds (information-time bounds) to characterize our capability to control the system when subject to arbitrary sources of noise. Moreover, since the presence of an external noise field induces open quantum system dynamics, we also show that the results provided by the information-time bounds are in very good agreement with the Kofman-Kurizki universal formula describing decoherence processes. Finally, we numerically test the scaling of the control accuracy as a function of the noise parameters, by means of the dressed chopped random basis (dCRAB) algorithm for quantum optimal control.

Shallow nitrogen-vacancy (NV) centers are promising candidates for high-precision sensing applications; these defects, when positioned a few nanometers below the surface, provide an atomic-scale resolution along with substantial sensitivity. However, the dangling bonds and impurities on the diamond surface result in a complex environment which reduces the sensitivity and is unique to each shallow NV center. To avoid the environment's detrimental effect, we apply feedback-based quantum optimal control. We first show how a direct search can improve the initialization and readout process. In a second step, we optimize microwave pulses for pulsed optically detected magnetic resonance (ODMR) and Ramsey measurements. Throughout the sensitivity optimizations, we focus on robustness against errors in the control field amplitude. This feature not only protects the protocols' sensitivity from drifts but also enlarges the sensing volume. The resulting ODMR measurements produce sensitivities below 1 mu T Hz(-1/2) for an 83% decrease in control power, increasing the robustness by approximately one third. The optimized Ramsey measurements produce sensitivities below 100 nT Hz(-1/2) giving a twofold sensitivity improvement. Being on par with typical sensitivities obtained via single NV magnetometry, the complementing robustness of the presented optimization strategy may provide an advantage for other NV-based applications.

The chopped random basis (CRAB) ansatz for quantum optimal control has been proven to be a versatile tool to enable quantum technology applications such as quantum computing, quantum simulation, quantum sensing, and quantum communication. Its capability to encompass experimental constraints-while maintaining an access to the usually trap-free control landscape-and to switch from open-loop to closed-loop optimization (including with remote access-or RedCRAB) is contributing to the development of quantum technology on many different physical platforms. In this review article we present the development, the theoretical basis and the toolbox for this optimization algorithm, as well as an overview of the broad range of different theoretical and experimental applications that exploit this powerful technique.

Adequate therapeutic retinal laser irradiation needs to be adapted to the local absorption. This leads to time-consuming treatments as the laser power needs to be successively adjusted to avoid under- and overtreatment caused by too low or too high temperatures. Closed-loop control can overcome this burden by means of temperature measurements. To allow for model predictive control schemes, the current state and the spot-dependent absorption need to be estimated. In this paper, we thoroughly compare moving horizon estimator (MHE) and extended Kalman filter (EKF) designs for joint state and parameter estimation. We consider two different scenarios, the estimation of one or two unknown absorption coefficients. For one unknown parameter, both estimators perform very similar. For two unknown parameters, we found that the MHE benefits from active parameter constraints at the beginning of the estimation, whereas after a settling time both estimators perform again very similar as long as the parameters are inside the considered parameter bounds.

Transcription factor programming of pluripotent stem cells (PSCs) has emerged as an approach to generate human neurons for disease modeling. However, programming schemes produce a variety of cell types, and those neurons that are made often retain an immature phenotype, which limits their utility in modeling neuronal processes, including synaptic transmission. We report that combining NGN2 programming with SMAD and WNT inhibition generates human patterned induced neurons (hpiNs). Single-cell analyses showed that hpiN cultures contained cells along a developmental continuum, ranging from poorly differentiated neuronal progenitors to well-differentiated, excitatory glutamatergic neurons. The most differentiated neurons could be identified using a CAMK2A::GFP reporter gene and exhibited greater functionality, including NMDAR-mediated synaptic transmission. We conclude that utilizing single-cell and reporter gene approaches for selecting successfully programmed cells for study will greatly enhance the utility of hpiNs and other programmed neuronal populations in the modeling of nervous system disorders.

Transcription factor programming of pluripotent stem cells (PSCs) has emerged as an approach to generate human neurons for disease modeling. However, programming schemes produce a variety of cell types, and those neurons that are made often retain an immature phenotype, which limits their utility in modeling neuronal processes, including synaptic transmission. We report that combining NGN2 programming with SMAD and WNT inhibition generates human patterned induced neurons (hpiNs). Single-cell analyses showed that hpiN cultures contained cells along a developmental continuum, ranging from poorly differentiated neuronal progenitors to well-differentiated, excitatory glutamatergic neurons. The most differentiated neurons could be identified using a CAMK2A::GFP reporter gene and exhibited greater functionality, including NMDAR-mediated synaptic transmission. We conclude that utilizing single-cell and reporter gene approaches for selecting successfully programmed cells for study will greatly enhance the utility of hpiNs and other programmed neuronal populations in the modeling of nervous system disorders.

Transcription factor programming of pluripotent stem cells (PSCs) has emerged as an approach to generate human neurons for disease modeling. However, programming schemes produce a variety of cell types, and those neurons that are made often retain an immature phenotype, which limits their utility in modeling neuronal processes, including synaptic transmission. We report that combining NGN2 programming with SMAD and WNT inhibition generates human patterned induced neurons (hpiNs). Single-cell analyses showed that hpiN cultures contained cells along a developmental continuum, ranging from poorly differentiated neuronal progenitors to well-differentiated, excitatory glutamatergic neurons. The most differentiated neurons could be identified using a CAMK2A::GFP reporter gene and exhibited greater functionality, including NMDAR-mediated synaptic transmission. We conclude that utilizing single-cell and reporter gene approaches for selecting successfully programmed cells for study will greatly enhance the utility of hpiNs and other programmed neuronal populations in the modeling of nervous system disorders.

Transcription factor programming of pluripotent stem cells (PSCs) has emerged as an approach to generate human neurons for disease modeling. However, programming schemes produce a variety of cell types, and those neurons that are made often retain an immature phenotype, which limits their utility in modeling neuronal processes, including synaptic transmission. We report that combining NGN2 programming with SMAD and WNT inhibition generates human patterned induced neurons (hpiNs). Single-cell analyses showed that hpiN cultures contained cells along a developmental continuum, ranging from poorly differentiated neuronal progenitors to well-differentiated, excitatory glutamatergic neurons. The most differentiated neurons could be identified using a CAMK2A::GFP reporter gene and exhibited greater functionality, including NMDAR-mediated synaptic transmission. We conclude that utilizing single-cell and reporter gene approaches for selecting successfully programmed cells for study will greatly enhance the utility of hpiNs and other programmed neuronal populations in the modeling of nervous system disorders.

Transcription factor programming of pluripotent stem cells (PSCs) has emerged as an approach to generate human neurons for disease modeling. However, programming schemes produce a variety of cell types, and those neurons that are made often retain an immature phenotype, which limits their utility in modeling neuronal processes, including synaptic transmission. We report that combining NGN2 programming with SMAD and WNT inhibition generates human patterned induced neurons (hpiNs). Single-cell analyses showed that hpiN cultures contained cells along a developmental continuum, ranging from poorly differentiated neuronal progenitors to well-differentiated, excitatory glutamatergic neurons. The most differentiated neurons could be identified using a CAMK2A::GFP reporter gene and exhibited greater functionality, including NMDAR-mediated synaptic transmission. We conclude that utilizing single-cell and reporter gene approaches for selecting successfully programmed cells for study will greatly enhance the utility of hpiNs and other programmed neuronal populations in the modeling of nervous system disorders. Nehme et al. combine two strong neuralizing factors (transcription factor programming and small molecule patterning) to generate human excitatory neurons from stem cells. They further undertake single-cell and reporter gene approaches to select highly differentiated neurons with increased functionality, augmenting their utility in the modeling of nervous system disorders.

Transcription factor programming of pluripotent stem cells (PSCs) has emerged as an approach to generate human neurons for disease modeling. However, programming schemes produce a variety of cell types, and those neurons that are made often retain an immature phenotype, which limits their utility in modeling neuronal processes, including synaptic transmission. We report that combining NGN2 programming with SMAD and WNT inhibition generates human patterned induced neurons (hpiNs). Single-cell analyses showed that hpiN cultures contained cells along a developmental continuum, ranging from poorly differentiated neuronal progenitors to well-differentiated, excitatory glutamatergic neurons. The most differentiated neurons could be identified using a CAMK2A::GFP reporter gene and exhibited greater functionality, including NMDAR-mediated synaptic transmission. We conclude that utilizing single-cell and reporter gene approaches for selecting successfully programmed cells for study will greatly enhance the utility of hpiNs and other programmed neuronal populations in the modeling of nervous system disorders.

Context Cultural landscapes evolve over time. However, the rate and direction of change might not be in line with societal needs and more information on the forces driving these changes are therefore needed. Objectives Filling the gap between single case studies and meta-analyses, we present a comparative study of landscape changes and their driving forces based in six regions across Europe conducted using a consistent method. Methods A LULC analysis based on historical and contemporary maps from the nineteenth and twentieth century was combined with oral history interviews to learn more about perceived landscape changes, and remembered driving forces. Land cover and landscape changes were analysed regarding change, conversions and processes. For all case study areas, narratives on mapped land cover change, perceived landscape changes and driving forces were compiled. Results Despite a very high diversity in extent, direction and rates of change, a few dominant processes and widespread factors driving the changes could be identified in the six case study areas, i.e. access and infrastructure, political shifts, labor market, technological innovations, and for the more recent period climate change. Conclusions Grasping peoples’ perception supplements the analyses of mapped land use and land cover changes and allows to address perceived landscape changes. The list of driving forces determined to be most relevant shows clear limits in predictability: Whereas changes triggered by infrastructural developments might be comparatively easy to model, political developments cannot be foreseen but might, nevertheless, leave major marks in the landscape.

Context Cultural landscapes evolve over time. However, the rate and direction of change might not be in line with societal needs and more information on the forces driving these changes are therefore needed. Objectives Filling the gap between single case studies and meta-analyses, we present a comparative study of landscape changes and their driving forces based in six regions across Europe conducted using a consistent method. Methods A LULC analysis based on historical and contemporary maps from the nineteenth and twentieth century was combined with oral history interviews to learn more about perceived landscape changes, and remembered driving forces. Land cover and landscape changes were analysed regarding change, conversions and processes. For all case study areas, narratives on mapped land cover change, perceived landscape changes and driving forces were compiled. Results Despite a very high diversity in extent, direction and rates of change, a few dominant processes and widespread factors driving the changes could be identified in the six case study areas, i.e. access and infrastructure, political shifts, labor market, technological innovations, and for the more recent period climate change. Conclusions Grasping peoples’ perception supplements the analyses of mapped land use and land cover changes and allows to address perceived landscape changes. The list of driving forces determined to be most relevant shows clear limits in predictability: Whereas changes triggered by infrastructural developments might be comparatively easy to model, political developments cannot be foreseen but might, nevertheless, leave major marks in the landscape.

Context Cultural landscapes evolve over time. However, the rate and direction of change might not be in line with societal needs and more information on the forces driving these changes are therefore needed. Objectives Filling the gap between single case studies and meta-analyses, we present a comparative study of landscape changes and their driving forces based in six regions across Europe conducted using a consistent method. Methods A LULC analysis based on historical and contemporary maps from the nineteenth and twentieth century was combined with oral history interviews to learn more about perceived landscape changes, and remembered driving forces. Land cover and landscape changes were analysed regarding change, conversions and processes. For all case study areas, narratives on mapped land cover change, perceived landscape changes and driving forces were compiled. Results Despite a very high diversity in extent, direction and rates of change, a few dominant processes and widespread factors driving the changes could be identified in the six case study areas, i.e. access and infrastructure, political shifts, labor market, technological innovations, and for the more recent period climate change. Conclusions Grasping peoples’ perception supplements the analyses of mapped land use and land cover changes and allows to address perceived landscape changes. The list of driving forces determined to be most relevant shows clear limits in predictability: Whereas changes triggered by infrastructural developments might be comparatively easy to model, political developments cannot be foreseen but might, nevertheless, leave major marks in the landscape.

Context Cultural landscapes evolve over time. However, the rate and direction of change might not be in line with societal needs and more information on the forces driving these changes are therefore needed. Objectives Filling the gap between single case studies and meta-analyses, we present a comparative study of landscape changes and their driving forces based in six regions across Europe conducted using a consistent method. Methods A LULC analysis based on historical and contemporary maps from the nineteenth and twentieth century was combined with oral history interviews to learn more about perceived landscape changes, and remembered driving forces. Land cover and landscape changes were analysed regarding change, conversions and processes. For all case study areas, narratives on mapped land cover change, perceived landscape changes and driving forces were compiled. Results Despite a very high diversity in extent, direction and rates of change, a few dominant processes and widespread factors driving the changes could be identified in the six case study areas, i.e. access and infrastructure, political shifts, labor market, technological innovations, and for the more recent period climate change. Conclusions Grasping peoples’ perception supplements the analyses of mapped land use and land cover changes and allows to address perceived landscape changes. The list of driving forces determined to be most relevant shows clear limits in predictability: Whereas changes triggered by infrastructural developments might be comparatively easy to model, political developments cannot be foreseen but might, nevertheless, leave major marks in the landscape.

Context Cultural landscapes evolve over time. However, the rate and direction of change might not be in line with societal needs and more information on the forces driving these changes are therefore needed. Objectives Filling the gap between single case studies and meta-analyses, we present a comparative study of landscape changes and their driving forces based in six regions across Europe conducted using a consistent method. Methods A LULC analysis based on historical and contemporary maps from the nineteenth and twentieth century was combined with oral history interviews to learn more about perceived landscape changes, and remembered driving forces. Land cover and landscape changes were analysed regarding change, conversions and processes. For all case study areas, narratives on mapped land cover change, perceived landscape changes and driving forces were compiled. Results Despite a very high diversity in extent, direction and rates of change, a few dominant processes and widespread factors driving the changes could be identified in the six case study areas, i.e. access and infrastructure, political shifts, labor market, technological innovations, and for the more recent period climate change. Conclusions Grasping peoples’ perception supplements the analyses of mapped land use and land cover changes and allows to address perceived landscape changes. The list of driving forces determined to be most relevant shows clear limits in predictability: Whereas changes triggered by infrastructural developments might be comparatively easy to model, political developments cannot be foreseen but might, nevertheless, leave major marks in the landscape.

Optimal control theory is a powerful tool for improving figures of merit in quantum information tasks. Finding the solution to any optimal control problem via numerical optimization depends crucially on the choice of the optimization functional. Here, we derive a functional that targets the full set of two-qubit perfect entanglers, gates capable of creating a maximally entangled state out of some initial product state. The functional depends on easily computable local invariants and unequivocally determines whether a gate is a perfect entangler. Optimization with our functional is most useful if the two-qubit dynamics allows for the implementation of more than one perfect entangler. We discuss the reachable set of perfect entanglers for a generic Hamiltonian that corresponds to several quantum information platforms of current interest.

The difficulty of an optimization task in quantum information science depends on the proper mathematical expression of the physical target. Here we demonstrate the power of optimization functionals targeting an arbitrary perfect two-qubit entangler, which allow generation of a maximally entangled state from some initial product state. We show for two quantum information platforms of current interest, i.e., nitrogen vacancy centers in diamond and superconducting Josephson junctions, that an arbitrary perfect entangler can be reached faster and with higher fidelity than both specific two-qubit gates and local equivalence classes of two-qubit gates. Our results are obtained using two independent optimization approaches, underscoring the critical role of the optimization target.

The difficulty of an optimization task in quantum information science depends on the proper mathematical expression of the physical target. Here we demonstrate the power of optimization functionals targeting an arbitrary perfect two-qubit entangler, which allow generation of a maximally entangled state from some initial product state. We show for two quantum information platforms of current interest, i.e., nitrogen vacancy centers in diamond and superconducting Josephson junctions, that an arbitrary perfect entangler can be reached faster and with higher fidelity than both specific two-qubit gates and local equivalence classes of two-qubit gates. Our results are obtained using two independent optimization approaches, underscoring the critical role of the optimization target.

Optimal control theory is a powerful tool for improving figures of merit in quantum information tasks. Finding the solution to any optimal control problem via numerical optimization depends crucially on the choice of the optimization functional. Here, we derive a functional that targets the full set of two-qubit perfect entanglers, gates capable of creating a maximally entangled state out of some initial product state. The functional depends on easily computable local invariants and unequivocally determines whether a gate is a perfect entangler. Optimization with our functional is most useful if the two-qubit dynamics allows for the implementation of more than one perfect entangler. We discuss the reachable set of perfect entanglers for a generic Hamiltonian that corresponds to several quantum information platforms of current interest.

To investigate the roles of tumor necrosis factor (TNF) and lymphotoxin (LT)-α in the development and function of the immune system, the Tnf and Ltα genes were simultaneously inactivated in mice by homologous recombination. These mutant mice are highly susceptible to Listeria monocytogenes infection and resistant to endotoxic shock induced by the combined administration of D-galactosamine (D-GaIN) and lipopolysaccharide (LPS). Their splenic microarchitecture is disorganized, characterized by the loss of the clearly defined marginal zone, ill defined T and B cell areas, and absence of MAdCAM-1 and reduced ICAM-1, VCAM-1 and Mac-1 expression. They are devoid of peripheral lymph nodes and Peyer's patches, and show a strong reduction of lgA+ plasma cells in the intestinal lamina propria. The alymphoplasia is accompanied by a marked B lymphocytosis and reduced basal Ig levels. Ig depositions in the renal glomerulus and a strong up-regulation of MHC class I antigen expression on endothelial cells of different tissues are observed. The primary humoral immune response towards sheep red blood cells reveals a defective IgG isotype switch, while that against vescicular stomatitis virus is normal. The cytotoxic T cell responses are attenuated, although still effective, against vaccinia, lymphocytic choriomeningitis virus (LCMV-ARM) and LCMV-WE. In conclusion, the combined inactivation of Tnf and Ltα confirms their essential role in the normal development and function of the immune system

Background Since May 2022, increasing numbers of monkeypox virus (MPXV) infections have been reported from across Europe and North America. Studies, mainly from Africa, have suggested a higher risk for severe MPXV cases in people living with HIV. Methods This was a retrospective study of all confirmed MPXV infections observed in the participating centres since 19 May 2022. We conducted a chart review to evaluate clinical characteristics, comorbidities, and coinfections, including HIV, viral hepatitis, and sexually transmitted infections (STIs). Results By 30 June 2022, a total of 546 MPXV infections were reported from 42 German centres. All patients were men who have sex with men (MSM), of whom 256 (46.9%) were living with HIV, mostly with a preserved immune system and with viral suppression. In total, 232 (42.5%) MSM were also taking HIV pre-exposure prophylaxis (PrEP) and 58 (10.6%) MSM had no known HIV infection or PrEP use. The median age was 39 years (range 20-67), and comorbidities were rare. However, 52.4% and 29.4% of all patients had been diagnosed with at least one STI within the last 6 months or within the last 4 weeks, respectively. The most frequent localizations of MPXV infection were genital (49.9%) and anal (47.9%), whereas fever (53.2%) and lymphadenopathy (42.6%) were the most frequent general symptoms. The hospitalization rate was low (4.0%), and no fatal course was observed. The clinical picture showed no apparent differences between MSM with or without HIV. Conclusions In this preliminary cohort analysis from a current large outbreak among MSM in Germany, the clinical picture of MPXV infection did not differ between MSM with and without HIV infection. Severe courses were rare and hospitalization rates were low. However, most patients were relatively healthy, and only a few people living with HIV were viremic or severely immunosuppressed.