Recent advances in text-to-image diffusion models spurred research on personalization, i.e., a customized image synthesis, of subjects within reference images. Although existing personalization methods are able to alter the subjects’ positions or to personalize multiple subjects simultaneously, they often struggle to modify the behaviors of subjects or their dynamic interactions. The difficulty is attributable to overfitting to reference images, which worsens if only a single reference image is available. We propose DynASyn, an effective multi-subject personalization from a single reference image addressing these challenges. DynASyn preserves the subject identity in the personalization process by aligning concept-based priors with subject appearances and actions. This is achieved by regularizing the attention maps between the subject token and images through concept-based priors. In addition, we propose concept-based prompt-and-image augmentation for an enhanced trade-off between identity preservation and action diversity. We adopt an SDE-based editing guided by augmented prompts to generate diverse appearances and actions while maintaining identity consistency in the augmented images. Experiments show that DynASyn is capable of synthesizing highly realistic images of subjects with novel contexts and dynamic interactions with the surroundings, and outperforms baseline methods in both quantitative and qualitative aspects.

Inherited retinal diseases (IRDs) are genetic disorders that cause progressive deterioration of the photoreceptors associated with vision loss or blindness. Retinitis pigmentosa (RP) is a rare hereditary ophthalmic disease that initially causes night blindness owing to continuous retinal pigment deterioration. A computer-aided diagnosis (CAD)-based RP diagnosis solution by pigment sign detection can help ophthalmologists to analyze and treat the disease timely. At present, most of the research addresses retinal disease CAD using expensive optical coherence tomography (OCT); however, fundus imaging-based solutions are quick, convenient, and inexpensive for massive screening. This study proposes two convolutional neural networks (CNNs)-based segmentation that combines multiscale features by spatial information fusion: a single spatial fusion network (SSF-Net) and a triplet spatial fusion network (TSF-Net). SSF-Net fuses four multiscale spatial information streams. TSF-Net exploits triplet spatial information fusion by early, intermediate, and late fusion to ensure the fine segmentation of retinal pigment signs without preprocessing. TSF-Net creates a valuable difference in performance over SSF-Net. To evaluate SSF-Net and TSF-Net, the open dataset, named Retinal Images for Pigment Signs is utilized with 4-fold cross-validation. The experiment results confirm that SSF-Net and TSF-Net demonstrate superior performance compared to the state-of-the-art methods for the screening and analysis of RP disease.

Robotic manipulators require contact force sensing capabilities to sense the contact force between the manipulator and an object. Specifically, for humans and robots to coexist in the work environment, the robot must be able to detect an external force applied by a human. This study presents a new intrinsic force sensing method for robot manipulators that can obtain accurate information of the external force applied by a human during human–robot interaction. The method employs a robot cover, which is typically utilized in robot manipulators. Unlike conventional force sensing methods, a six-axis force/torque sensor is placed between the cover and the link of the robot manipulator. As a result, the proposed method provides information of the three-axis contact force applied to the cover surface and its contact location. Therefore, the cover itself becomes a sensorized cover based on the intrinsic force sensing method. To evaluate its sensing performance, the sensorized cover is experimentally validated using reference sensors. Finally, an experiment is performed in which the robot successfully recognizes letters written on the cover, indicating a high level of contact force sensing performance.

Many optimization problems in recent engineering are complex and high-dimensional problems, a so-called Large-Scale Global Optimization (LSGO) problem, due to the increasing requirements for multidisciplinary approach. This paper proposes a novel Bare Bones Particle Swarm Optimization (BBPSO) algorithm to solve LSGO problems. The BBPSO is a variant of a Particle Swarm Optimization (PSO) and is based on Gaussian distribution. The BBPSO does not consider the selection of controllable parameters of the PSO and is a simple but powerful optimizer. This algorithm, however, is vulnerable to LSGO problems. This study has improved its performance for LSGO problems by combining the heterogeneous cooperation based on the information exchange between particles and the Gaussian jump strategy to avoid local optima. The CEC'2015 Special Session on Large-Scale Global Optimization has given 15 benchmark problems to provide convenience and flexibility for comparing various optimization algorithms specifically designed for large-scale global optimization. Simulations performed with those benchmark problems have verified the performance of the proposed optimizer and compared with the reference algorithm DECC-G of the CEC'2015 special session on LSGO.

All of industrial robots need the fixed-time control property commonly. The EtherCAT is a highly flexible network protocol that is developing at a rapid rate. It is extention of conventional ethernet with a high speed communication and simple wiring characteristic. With the ethercat, the implementation of the industrial robot is more easier. Here, the implementation of the industrial robot with the function of the direct teaching is described.