This study explores how Low-Rank Adaptation (LoRA) fine-tuning, guided by human aesthetic evaluations, can enhance the outputs of generative AI models in tangible product design, using lamp design as a case study. By integrating human feedback into the AI model, we aim to improve both the desirability and aesthetic appeal of the generated designs. Comprehensive experiments were conducted, starting with prompt optimization techniques and focusing on LoRA fine-tuning of the Stable Diffusion model. Additionally, methods to convert AI-generated designs into tangible products through 3D realization using 3D printing technologies were investigated. The results indicate that LoRA fine-tuning effectively aligns AI-generated designs with human aesthetic preferences, leading to significant improvements in desirability and aesthetic appeal scores. These findings highlight the potential of human-AI collaboration in tangible product design and provide valuable insights into integrating human feedback into AI design processes.

Terrestrial illumination undergoes continuous spectral, angular, spatial, and temporal changes throughout the day, influenced by diurnal cycles and atmospheric conditions such as haze. These variations in light exposure impact human physiology and behavior, particularly in providing "zeitgebers" (time givers) for biological rhythms. We analyzed spectral light-field data collected outdoors from dawn to dusk on four days: two in Delft (sunny and cloudy) and two overcast (uniformly cloud covered) days in Newcastle. By decomposing the light field into diffuse and directional components, we identified differences in spectral composition between these for all conditions, with overcast days showing reduced variability due to increased light scattering. Our study also explored the physiological implications for circadian regulation via melanopsin and other photoreceptors. We found that α-opic illumination vectors varied with weather, their order aligning with the sequence of photoreceptor spectral peak sensitivities—from S cones, to ipRGCs, rods, M cones, and L cones—especially under sunny and cloudy skies. Analysis of hazy versus clear images revealed that haze shifts chromaticity towards blue, potentially enhancing melanopic efficiency. Although these fluctuations are large, it is plausible that they do not impact biological rhythms to the same extent as illumination variations at dawn and dusk, ensuring that the latter remain the primary drivers of circadian rhythm regulation.