Abstract Achieving physically consistent image editing remains a significant challenge in computer vision. Existing image editing methods typically rely on neural networks, which struggle to accurately handle shadows and refractions. Conversely, physics-based inverse rendering often requires multi-view optimization, limiting its practicality in single-image scenarios. In this paper, we propose Materialist , a neural-initialized physically based rendering pipeline for single-image inverse rendering. Unlike previous hybrid methods that use physics to guide neural generation, our method leverages neural networks to predict initial material properties, which are then rigorously optimized via progressive differentiable rendering. Our approach enables a range of applications, including material editing, object insertion, and relighting, while also introducing an effective method for editing material transparency via ray-traced refraction without requiring full scene geometry. Furthermore, our envmap estimation method also achieves competitive performance, further enhancing the accuracy of image editing task. Experiments demonstrate strong performance across synthetic and real-world datasets, excelling even on challenging out-of-domain images.

Abstract We present a method for capturing the BSSRDF (bidirectional scattering‐surface reflectance distribution function) of arbitrary geometry with a neural network. We demonstrate how a compact neural network can represent the full 8‐dimensional light transport within an object including heterogeneous scattering. We develop an efficient rendering method using importance sampling that is able to render complex translucent objects under arbitrary lighting. Our method can also leverage the common planar half‐space assumption, which allows it to represent one BSSRDF model that can be used across a variety of geometries. Our results demonstrate that we can render heterogeneous translucent objects under arbitrary lighting and obtain results that match the reference rendered using volumetric path tracing.

In the Semantic Web context, OWL ontologies represent the conceptualization of domains of interest while the corresponding assertional knowledge is given by RDF data referring to them. Because of its open, distributed, and collaborative nature, such knowledge can be incomplete, noisy, and sometimes inconsistent. By exploiting the evidence coming from the assertional data, we aim at discovering hidden knowledge patterns in the form of multi-relational association rules while taking advantage of the intensional knowledge available in ontological knowledge bases. An evolutionary search method applied to populated ontological knowledge bases is proposed for finding rules with a high inductive power. The proposed method, EDMAR, uses problem-aware genetic operators, echoing the refinement operators of ILP, and takes the intensional knowledge into account, which allows it to restrict and guide the search. Discovered rules are coded in SWRL, and as such they can be straightforwardly integrated within the ontology, thus enriching its expressive power and augmenting the assertional knowledge that can be derived. Additionally, discovered rules may also suggest new axioms to be added to the ontology. We performed experiments on publicly available ontologies, validating the performances of our approach and comparing them with the main state-of-the-art systems.