上海工艺美术职业学院(Shanghai Art & Design Academy )是上海市唯一一所独立设置的艺术设计类高职院校,是上海市唯一承担国家教育体制改革试点的高职院校,是上海市唯一一所入选“中国特色高水平高职学校和专业建设计划”的高职院校,是中国非物质文化遗产传承人群研修基地、国家示范性高等职业院校、国家优质专科高等职业院校、全国毕业生就业典型经验高校、全国创新创业典型经验高校、上海市首批深化创新创业教育改革示范高校、上海市“一流高职”建设院校。学院由上海市工艺美术学校(建于1960年)与上海第二轻工业职工大学(建于1979年)于2003年联合组建。2010年,学院通过教育部、财政部国家示范性高等职业院校验收。2019年,学院被教育部、财政部列入中国特色高水平高职学校和专业建设计划建设单位。根据2017年9月学校官网信息显示,学院占地面积121299平方米,教学科研仪器设备资产总值5057万元;设有6个二级学院,开设高职(专科)专业18个;有专任教师220余人,在校生4096人。
This article proposes an integrated optimization method for full process design and production based on intelligent manufacturing to address the core contradictions of multiple varieties, small batches, high customization of cultural and creative products, and the disconnection from traditional design and production processes, high costs, long cycles, and unstable quality. Using digital twins, AIGC, CPS (Cyber Physical Systems), and flexible manufacturing as the core technologies, an integrated technical framework is constructed that includes intelligent requirement analysis, generative creative design, virtual verification, automated process planning, dynamic production scheduling, and a data -driven closed-loop for quality lifecycle control.
Multisensory exhibition design requires coordination between spatial layout and sound provision. Understanding how spatial configuration relates to visual exploration and experienced soundscapes is essential to that coordination. This study combined visibility graph analysis (VGA) with synchronous eye-tracking and functional near-infrared spectroscopy (fNIRS) from 36 participants and scene-specific questionnaires during interactive panoramic exhibition viewing. High-integration scenes showed longer exported coordinate paths, more regional switching and higher soundscape ratings, whereas medium-integration scenes showed longer fixations. Oxygenated haemoglobin (HbO) coefficients in three channels were lower under high than low integration. Exhibition-adapted eventfulness scores were higher during the second viewing with added sound, most strongly in high-integration scenes. Within the high-integration stratum, a joint behavioural–haemodynamic model explained 68.1% of eventfulness score variance, with a Q2 of 0.583 under cross-validation with one participant held out; visual behaviour provided the principal incremental contribution. The findings link visibility structure to distinct exploration and appraisal profiles, informing the coordination of exhibition layouts and sound.
Traditional clothing design relies on manual drawing or computer graphics software to complete the design. Machine learning can automatically extract design elements and style features through learning and analyzing a large amount of data, providing intelligent design suggestions and solutions for designers, thereby improving design efficiency and quality. The research adopts depth camera sensor technology to obtain the three-dimensional position information of the human body and uses machine learning algorithms to process and analyze the depth data. In terms of image segmentation, by integrating fuzzy clustering algorithms and deep learning algorithms, precise segmentation and recognition of clothing graphics can be achieved. Build a knowledge base of clothing experts through machine learning algorithms to provide intelligent design suggestions and solutions for designers. In the experiment, the segmentation effects of different algorithms were evaluated. The experimental results show that the image segmentation algorithm based on machine learning performs well in terms of segmentation accuracy and robustness, especially when dealing with complex scenes and noise, it can achieve more accurate segmentation effects. Compared with traditional methods, machine learning algorithms can significantly improve design efficiency, shorten the design cycle, and provide more personalized design suggestions. The RSNI_FCM algorithm performs best in terms of image segmentation quality and noise robustness, and can better balance noise robustness and detail retention ability. By combining depth camera sensors with image segmentation algorithms, this system can achieve more accurate human body measurement and clothing positioning, helping designers understand human body structure and needs more precisely. Machine learning algorithms can also enhance design efficiency and quality, reducing the time and cost required by traditional fitting models through virtual fitting and real-time adjustments.
Indoor visual positioning in complex indoor spaces requires reliable classification and detection under changing lighting, occlusion, and multi-category interference. To enhance recognition accuracy and real-time performance, a visual orientation system, integrating multi-feature fusion, incremental learning, and a regional proposal detection mechanism, is developed. Deep convolutional, gradient structural features, and texture descriptors are fused to enhance scenario representation, while incremental learning enables category expansion without degrading existing recognition ability. The regional proposal detection module improves target localization and boundary fitting in cluttered indoor layouts. Experimental results on the SUN RGB-D dataset show that the classification model achieves an accuracy of 0.92, a precision of 0.91, a recall of 0.90, an F1-score of 0.91, and a frame rate of 48 frames per second. The detection model achieves an accuracy of 0.94, a precision of 0.92, an Intersection over Union (IoU) score of 0.83, and a processing speed of 38 frames per second. These findings demonstrate that the proposed model achieves a strong balance between recognition performance and operational efficiency, offering practical support for indoor navigation and spatial orientation in dynamic environments.
In recent years, Artificial Intelligence (AI) has developed as a cutting-edge technology in the field of creative and cultural product design playing an important role in improving personalization and preserving cultural heritage. The existing recommendation systems which unbale to obtain complex cultural semantics, aesthetic preferences and emotional factors. To overcome these challenges, an advanced Explainable Transformer-Driven Generative Recommendation (ETDGR) framework for the personalized design of creative and cultural products. Moreover, the proposed model incorporates different intelligent mechanisms for improving design creativity and recommendation accuracy. Further, for extracting fine-grained textual and behavioral user preferences from the product reviews a transformer-based module is employed. Correspondingly, Graph Neural Network (GNN) is used for modelling non-linear and complex relationships among products and users as well as allowing efficient relational learning. Subsequently, for improving user trust and transparency an Explainable Artificial Intelligence (XAI) module integrates attention mechanisms and SHAP. Additionally, for creating novel and culturally meaningful product designs a Generative AI component regarding to diffusion models or GAN is presented. Furthermore, for effectively handle the uncertainty in user responses a Fuzzy Decision Support System (FDSS) is utilized and also this has the ability to transform ambiguity values into structured decision criteria. The experimental results displayed that the proposed ETDGR model substantially outperforms the traditional IAFSA-BP method in terms of precision of 96.6 and f1-score of 95.7.