Special skills are required in portrait painting, such as imagining geometric structures and facial detail for final portrait designs. This makes it a difficult task for users, especially novices without prior artistic training, to draw freehand portraits with high-quality details. In this paper, we propose dualFace, a portrait drawing interface to assist users with different levels of drawing skills to complete recognizable and authentic face sketches. Inspired by traditional artist workflows for portrait drawing, dualFace gives two-stages of drawing assistance to provide global and local visual guidance. The former helps users draw contour lines for portraits (i.e., geometric structure), and the latter helps users draw details of facial parts, which conform to the user-drawn contour lines. In the global guidance stage, the user draws several contour lines, and dualFace then searches for several relevant images from an internal database and displays the suggested face contour lines on the background of the canvas. In the local guidance stage, we synthesize detailed portrait images with a deep generative model from user-drawn contour lines, and then use the synthesized results as detailed drawing guidance. We conducted a user study to verify the effectiveness of dualFace, which confirms that dualFace significantly helps users to produce a detailed portrait sketch.

We conducted a one-year prospective, multicenter study to assess clinical outcomes after implantation of segmented, rotationally asymmetric multifocal intraocular lenses (IOLs) with +1.5 diopters (D) near addition. In this phase III clinical trial, 120 eyes of 65 patients undergoing phacoemulsification and implantation of Lentis Comfort LS-313 MF15 (Oculentis GmbH) were included. The ophthalmological examinations were performed before and 1 day, 1 week, 1, 3, 6, 9, and 12 months after surgery. Uncorrected (UDVA) and corrected (CDVA) distance visual acuity, uncorrected (UIVA) and distance-corrected (DCIVA) intermediate visual acuity at 70 cm, and uncorrected (UNVA) and distance-corrected (DCNVA) near visual acuity at 30 cm were measured. A defocus curve was obtained and patients were asked about the severity of photic phenomena. Postoperative distance and intermediate visual acuity was excellent, with UDVA, CDVA, UIVA, and DCIVA of approximately 20/20, 20/16, 20/25, 20/25 were attained, respectively. The level of near visual acuity was lower; UNVA and DCNVA remained at around 20/60 and 20/70, respectively. The defocus curve indicated that postoperative uncorrected visual acuity of 20/25 and 20/40 was obtained at as close as 67 cm and 50 cm, respectively. Contrast sensitivity was within the normal range, with a minimal level of subjective symptoms and high patient satisfaction. The rotationally asymmetric multifocal IOLs with +1.5 D near addition provided excellent distance and intermediate vision, but near vision was not enough for reading small prints. Contrast sensitivity was high, with very low incidences of photic phenomena and a high level of patient satisfaction.

At present, buildings are increasingly being designed with transparent materials, with glass paneling being especially popular as an installation material due to its architectural allure. However, its major drawback is admitting impractical amounts of sunlight into interior spaces. Office buildings with excessive sunlight in indoor areas lead to worker inefficiency. This article studied kinetic façades as means to provide suitable sunlight for interior spaces, integrated with a triple-identity DNA structure, photosynthetic behavior, and the twist, which was divided into generation and evaluation. The generating phase first used an evolutionary engine to produce potential strip patterns. The kinetic façade was subsequently evaluated using the Climate Studio software to validate daylight admission in an indoor space with Leadership in Energy and Environmental Design (LEED) version 4.1 criteria. To analyze the kinetic façade system, the building envelope was divided into four types: glass panel, static façade, rotating façade (the kinetic façade, version 1); an existing kinetic façade that is commonly seen in the market, and twisting façade (the kinetic façade, version 2); the kinetic façade that uses the process to invent the new identity of the façade. In addition, for both the rotating façade and twisting façade, the degrees of simulation were 20, 50, 80, and 100 degrees, in order to ascertain the potential for both façades to the same degree. Comparing all façades receiving the daylight factor (DF) into the space with more or less sunlight resulted in a decreasing order of potential, as follows: entirely glass façade, twisting façade (the kinetic façade, version 2), rotating façade (the kinetic façade, version 1), and static façade. By receiving the daylight factor (DF), the façade moderately and beneficially filtered appropriate amounts of daylight into the working space. The daylight simulation results indicated that the newly designed kinetic façade (version 2) had more potential than other building envelope types in terms of filtering beneficial daylight in indoor areas. This article also experimented with the kinetic façade prototype in an actual situation to test conditional environmental potential. The twisting façade (the kinetic façade, version 2) was explored in the building envelope with varied adaptability to provide sunlight and for private-to-public, public-to-private, or semi-public working areas.

Synthesizing facial images from monochromatic sketches is one of the most fundamental tasks in the field of image-to-image translation. However, it is still challenging to teach model high-dimensional face features, such as geometry and color, and to the characteristics of input sketches, which should be considered simultaneously. Existing methods often use sketches as indirect inputs (or as auxiliary inputs) to guide models, resulting in the loss of sketch features or in alterations to geometry information. In this paper, we introduce a Sketch-Guided Latent Diffusion Model (SGLDM), an LDM-based network architecture trained on the paired sketch–face dataset. We apply a Multi-Auto-Encoder (AE) to encode the different input sketches from the various regions of a face from the pixel space into a feature map in the latent space, enabling us to reduce the dimensions of the sketch input while preserving the geometry-related information of the local face details. We build a sketch-face paired dataset based on an existing method XDoG[30] xdog and Sketch Simplification [24],[25] that extracts the edge map from an image. We then introduce a Stochastic Region Abstraction (SRA), an approach to augmenting our dataset to improve the robustness of the SGLDM to handle arbitrarily abstract sketch inputs. The evaluation study shows that the SGLDM can synthesize high-quality face images with different expressions, facial accessories, and hairstyles from various sketches having different abstraction levels, and the code and model have been released on the project page. https://puckikk1202.github.io/difffacesketch2023/.