We use fused deposition modeling (FDM) 3D printing as a case study of how manufacturing robots can use imperfect AI to acquire process expertise. In FDM, print configuration strongly affects output quality. Yet, novice users typically rely on default configurations, trial-and-error, or recommendations from generic AI models (e.g., ChatGPT). These strategies can produce complete prints, but they do not reliably meet specific objectives. Experts iteratively tune print configurations using evidence from prior prints. We present a modular closed-loop approach that treats an LLM as a source of tuning expertise. We embed this source of expertise within a Bayesian optimization loop. An approximate evaluator scores each print configuration and returns structured diagnostics, which the LLM uses to propose natural-language adjustments that are compiled into machine-actionable guidance for optimization. On 100 Thingi10k parts, our LLM-guided loop achieves the best configuration on 78% objects with 0% likely-to-fail cases, while single-shot AI model recommendations are rarely best and exhibit 15% likely-to-fail cases. These results suggest that LLMs provide more value as constrained decision modules in evidence-driven optimization loops than as end-to-end oracles for print configuration selection. We expect this result to extend to broader LLM-based robot programming.
Existing Image Restoration (IR) studies typically focus on task-specific or universal modes individually, relying on the mode selection of users and lacking the cooperation between multiple task-specific/universal restoration modes. This leads to insufficient interaction for unprofessional users and limits their restoration capability for complicated real-world applications. In this work, we present HybridAgent, intending to incorporate multiple restoration modes into a unified image restoration model and achieve intelligent and efficient user interaction through our proposed hybrid agents. Concretely, we propose the hybrid rule of fast, slow, and feedback restoration agents. Here, the slow restoration agent optimizes the powerful multimodal large language model (MLLM) with our proposed instruction-tuning dataset to identify degradations within images with ambiguous user prompts and invokes proper restoration tools accordingly. The fast restoration agent is designed based on a lightweight large language model (LLM) via in-context learning to understand the user prompts with simple and clear requirements, which can obviate the unnecessary time/resource costs of MLLM. Moreover, we introduce the mixed distortion removal mode for our HybridAgents, which is crucial but not concerned in previous agent-based works. It can effectively prevent the error propagation of step-by-step image restoration and largely improve the efficiency of the agent system. We validate the effectiveness of HybridAgent with both synthetic and real-world IR tasks.
A high order time stepping applied to spatial discretizations provided by the method of lines for hyperbolic conservations laws is presented. This procedure is related to the one proposed in Qiu and Shu (SIAM J Sci Comput 24(6):2185-2198, 2003) for numerically solving hyperbolic conservation laws. Both methods are based on the conversion of time derivatives to spatial derivatives through a Lax-Wendroff-type procedure, also known as Cauchy-Kovalevskaya process. The original approach in Qiu and Shu (2003) uses the exact expressions of the fluxes and their derivatives whereas the new procedure computes suitable finite difference approximations of them ensuring arbitrarily high order accuracy both in space and time as the original technique does, with a much simpler implementation and generically better performance, since only flux evaluations are required and no symbolic computations of flux derivatives are needed.
Yangliuqing woodblock prints, a cornerstone of China's intangible cultural heritage, are celebrated for their intricate designs and vibrant colors. However, preserving these traditional art forms while fostering innovation presents significant challenges. This study explores the DeepSeek + MidJourney approach to generating creative, themed Yangliuqing woodblock prints focused on the fight against COVID-19 and depicting joyous winners. Using Fréchet Inception Distance (FID) scores for evaluation, the method that combined DeepSeek-generated thematic prompts, MidJourney-generated thematic images, original Yangliuqing prints, and DeepSeek-generated key prompts in MidJourney-generated outputs achieved the lowest mean FID score (150.2) with minimal variability (σ = 4.9). Additionally, feedback from 62 participants, collected via questionnaires, confirmed that this hybrid approach produced the most representative results. Moreover, the questionnaire data revealed that participants demonstrated the highest willingness to promote traditional culture and the strongest interest in consuming the AI-generated images produced through this method. These findings underscore the effectiveness of an innovative approach that seamlessly blends traditional artistic elements with modern AI-driven creativity, ensuring both cultural preservation and contemporary relevance.
AbstractPyrolytic synergistic interactions, in which the production of pyrolyzates is enhanced or inhibited, commonly occur during the co-pyrolysis of different polymeric materials, such as plastics and biomass. Although these interactions can increase the yield of desired pyrolysis products under controlled degradation conditions, the desired compounds must be separated from complex pyrolyzates and further purified. To balance these dual effects, this study was aimed at examining pyrolytic synergistic interactions during slow heating co-pyrolysis of biodegradable plastics including polylactic acid (PLA) and poly(3-hydroxybutyrate-co-3-hydroxyhexaoate) (PHBH) and petroleum-based plastics including high-density polyethylene (HDPE), polypropylene (PP), and polystyrene (PS). Comprehensive investigations based on thermogravimetric analysis, pyrolysis–gas chromatography/mass spectrometry, and evolved gas analysis-mass spectrometry revealed that PLA and PHBH decompose at lower temperatures (273–378 °C) than HDPE, PP, and PS (386–499 °C), with each polymer undergoing independent decomposition without any pyrolytic interactions. Thus, the independent pyrolysis of biodegradable plastics, such as PLA and PHBH, with common plastics, such as HDPE, PP, and PS, can theoretically be realized through temperature control, enabling the selective recovery of their pyrolyzates in different temperature ranges. Thus, pyrolytic approaches can facilitate the treatment of mixed biodegradable and common plastics.
José Ramón Fernández Molina, José Ramón Hevia González
La intervención en el tramo de la calle Paraíso de la muralla de Oviedo ha permitido su recuperación y puesta en valor, dando a conocer ampliamente este Bien de Interés Cultural (declarado monumento histórico-artístico por Decreto de 3 de junio de 1931), indispensable para entender la propia conformación y evolución de la ciudad. Una ciudad cercada por un recinto murado a la que Alfonso II en el siglo IXtrasladó la corte de la monarquía asturiana, y desde la cual se iniciaría la primera peregrinación a Compostela para visitar el sepulcro recién descubierto del apóstol Santiago. Siglos más tarde, en plena Reconquista, Alfonso X ordenaría construir una nueva cerca. En el presente artículo se analiza la situación jurídica y patrimonial previa a la reciente intervención, las necesidades contractuales, así como la evolución de la fábrica, los tratamientos técnicos y las decisiones proyectuales adoptadas por el equipo pluridisciplinar comisionado, encabezado por el arquitecto José Ramón Fernández Molina.
Conservation and restoration of prints, Architectural drawing and design
Urban regeneration and industrial archaeology, dating back between the fourties and the fifties of the 20th century, has drawn attention to architecture, equipment and evidence of human manufacturing, widening the scope of Heritage Preservation in Europe and all over the world. Scholars detected an overall and quite similar development of the approach to industrial Heritage throughout time in the western world, from an early spontaneus interest, through an exploitation mostly based on economic purposes, towards a more recent preservative and inclusive attitude in regenerating buildings and sites, that has a strong support from the voluntary sector in UK. This paper reports the concept and main achievements of a team that, since the seventies, has being pioneering an interdisciplinary, cost-effective and yet preservative approach to abandoned production sites. Blending economic return and true attention to people and site specificity, innovation and campaigns for protection and proper reuse, this approach has been proving to be able to decline the different acceptations of sustainability and, as for conceptual and operational aspects, presents some interesting food for thought.
Conservation and restoration of prints, Architectural drawing and design
Tras la restitución del Generalife al Estado español en 1921, Benigno de la Vega Inclán (1858-1942), el marqués de la Vega-Inclán organizó un patronato independiente del de la Alhambra para atender la apertura del conjunto nazarí a los futuros visitantes. Eladio Laredo y Carranza (1864-1941) fue el arquitecto encargado de avanzar los primeros levantamientos, estudio de accesos, riegos y recuperación de las zonas degradadas, hasta 1925, en que se unifican las competencias en la figura del arquitecto Leopoldo Torres Balbás.
Conservation and restoration of prints, Architectural drawing and design
Yuyang Hu, Mauricio Delbracio, Peyman Milanfar
et al.
Image denoisers have been shown to be powerful priors for solving inverse problems in imaging. In this work, we introduce a generalization of these methods that allows any image restoration network to be used as an implicit prior. The proposed method uses priors specified by deep neural networks pre-trained as general restoration operators. The method provides a principled approach for adapting state-of-the-art restoration models for other inverse problems. Our theoretical result analyzes its convergence to a stationary point of a global functional associated with the restoration operator. Numerical results show that the method using a super-resolution prior achieves state-of-the-art performance both quantitatively and qualitatively. Overall, this work offers a step forward for solving inverse problems by enabling the use of powerful pre-trained restoration models as priors.
Achilles Bergne, Guido Baardink, Evripides G Loukaides
et al.
Mechanical metamaterials are structures designed to exhibit an exotic response, such as topological soft modes at a surface. Here we explore single-material 3D prints of these topological structures by translating a ball-and-spring model into a physical prototype. By uniaxially compressing the 3D-printed solid having marginal rigidity, we observe that the surfaces are consistently softer than the bulk. However, we also find that either of two opposite surfaces can be the softest, in contrast to the topologically robust predictions of the linear model. Finite-element simulations allow us to bridge this gap. We explore how the printing geometry and deformation amplitude could affect surface softness. For small strains, we find qualitative agreement with the ball-and-spring model but, surprisingly, nonlinear deformations can select which side is softest. Our work contextualizes the predictions of topological mechanics for real 3D materials and their potential for cushioning applications.
Face Restoration (FR) aims to restore High-Quality (HQ) faces from Low-Quality (LQ) input images, which is a domain-specific image restoration problem in the low-level computer vision area. The early face restoration methods mainly use statistical priors and degradation models, which are difficult to meet the requirements of real-world applications in practice. In recent years, face restoration has witnessed great progress after stepping into the deep learning era. However, there are few works to systematically study the deep learning based face restoration methods. Thus, in this paper, we provide a comprehensive survey of recent advances in deep learning techniques for face restoration. Specifically, we first summarize different problem formulations and analyze the characteristics of face images. Second, we discuss the challenges of face restoration. With regard to these challenges, we present a comprehensive review of recent FR methods, including prior-based methods and deep-learning methods. Then, we explore developed techniques in the task of FR covering network architectures, loss functions, and benchmark datasets. We also conduct a systematic benchmark evaluation on representative methods. Finally, we discuss the future directions including network designs, metrics, benchmark datasets, applications, etc. We also provide an open source repository for all the discussed methods, which is available at https://github.com/TaoWangzj/Awesome-Face-Restoration.
El antiguo asentamiento fortificado de Bam (Arg-e Bam o Ciudadela de Bam), un testimonio excelente del desarrollo de un enclave comercial en una región desértica de Asia, es un caso excepcional en el cual la ósmosis entre el entorno construido y la naturaleza es total e incluso se halla enriquecida por su aspecto cromático y su suave relieve resultante de la interacción entre materiales y factores atmosféricos. Tras el grave terremoto de 2013, Arg-e Bam representa también un ejemplo significativo de reconstrucción masiva y, al mismo tiempo, un modelo real para analizar el concepto de autenticidad del patrimonio.
Conservation and restoration of prints, Architectural drawing and design
Andorra is a small country nestled in the heart of the Pyrenees that has undergone a radical transformation over the past half century. From a centuries-old mountain culture based on subsistence agriculture, the country saw itself converted into a leading tourist destination in a very short period of time. This change has been accompanied by significantly high levels of immigration. The traditional society of the country, which had the family house at the centre of social life, soon found itself relegated along with much of its material evidence. In recent years, the country has started to look back at the past in an attempt to connect the new generations with their own history. With this end in mind, Andorra has been implementing a policy focussed on recovering its traditional legacy and encouraging the renovation of its vernacular architecture. The enactment of the Cultural Heritage Act in 2003 has played a fundamental role in these efforts. The interventions of recent years have given us a greater insight into traditional Andorran construction, the conservation problems associated with it as well as the merits of the solutions that have been adopted. In this article, we review the history of this legacy and the manner in which it is being restored.
Conservation and restoration of prints, Architectural drawing and design
Mercé Zazurca, Joan Ramon Rosell, Montserrat Bosch
et al.
Josep Puig i Cadafalch (Mataró 1867-Barcelona 1956), una de las figuras más polifacéticas de Cataluña, dejó un valioso legado, como es ejemplo la casa de veraneo de Argentona (Barcelona). Lamentablemente, esta casa fue deteriorándose hasta que en 2010 cayeron parte de las almenas de fachada arrastrando parcialmente la cubierta, lo que obligó a tomar una decisión provisional de estabilización y, posteriormente, a convocar un concurso de proyectos con el fin de recuperar un edificio emblemático y capital para entender la obra y el personaje de Puig i Cadafalch. En este artículo se presentan los trabajos de rehabilitación y recuperación de la cubierta como parte de un proyecto global que pretende convertir el edificio en un potente polo dinamizador de Argentona.
Conservation and restoration of prints, Architectural drawing and design
Three-dimensional concrete printing (3DCP) has gained a lot of popularity in recent years. According to many, 3DCP is set to revolutionize the construction industry: yielding unparalleled aesthetics, better quality control, lower cost, and a reduction of the construction time. In this paper, two finite element method (FEM) strategies are presented for simulating such 3D concrete printing processes. The aim of these models is to predict the structural behaviour during printing, while the concrete is still fresh, and estimate the optimal print speed and maximum overhang angle to avoid print failures. Both FE analyses involve solving multiple static implicit steps where sets of finite elements are added stepwise until failure. The main difference between the two methods is in the discretization of the 3D model. The first method uses voxelization to approximate the 3D shape, while the second approach starts from defining the toolpath and constructs finite elements by sweeping them along the path. A case study is presented to evaluate the effectiveness of both strategies. Both models are in good agreement with each other, and a comparable structural response is obtained. The model's limitations and future challenges are also discussed. Ultimately, the paper demonstrates how FEM-based models can effectively simulate complex prints and could give recommendations with regards to a better print strategy. These suggestions can be related to the maximum printing speed and overhang angle, but also the optimal layer height and thickness, the specific choice of the infill pattern, or by extension the mixture design. When print failures can be avoided, this methodology could save time, resources and overall cost. Future work will focus on the validation of these numerical models and comparing them to experimental data.
John You En Chan, Qifeng Ruan, Menghua Jiang
et al.
A light field print (LFP) displays three-dimensional (3D) information to the naked-eye observer under ambient white light illumination. Changing perspectives of a 3D image are seen by the observer from varying angles. However, LFPs appear pixelated due to limited resolution and misalignment between their lenses and colour pixels. A promising solution to create high-resolution LFPs is through the use of advanced nanofabrication techniques. Here, we use two-photon polymerization lithography as a one-step nanoscale 3D printer to directly fabricate LFPs out of transparent resin. This approach produces simultaneously high spatial resolution (29 - 45 μm) and high angular resolution (~ 1.6 °) images with smooth motion parallax across 15 {\times} 15 views. Notably, the smallest colour pixel consists of only a single nanopillar (~ 300 nm diameter). Our LFP signifies a step towards hyper-realistic 3D images that can be applied in print media and security tags for high-value goods.