🤖 AI Summary
This work addresses the challenge of co-optimizing architectural form generation and physical performance. We propose a novel Jenga-inspired modular stacking paradigm that integrates 6D pose estimation with wind-environment assessment. For the first time, open-set 6D pose estimation algorithms (e.g., Gen6D) are incorporated into physics-driven architectural design workflows, leveraging multi-view RGB inputs, real-time inference acceleration, and bidirectional coupling with wind simulation. The method achieves centimeter-level pose accuracy under severe occlusion and millisecond-scale response latency. Experimental results demonstrate substantial improvements in structural stability and wind-load adaptability, validating the feasibility of pose-guided morphogenesis. The study further identifies key limitations of current 6D pose estimators in architectural contexts—including occlusion robustness and cross-domain generalizability—and delineates algorithmic enhancement pathways explicitly tailored to closed-loop feedback from building physical performance metrics.
📝 Abstract
This paper includes a review of current state of the art 6d pose estimation methods, as well as a discussion of which pose estimation method should be used in two types of architectural design scenarios. Taking the latest pose estimation research Gen6d as an example, we make a qualitative assessment of the current openset methods in terms of application level, prediction speed, resistance to occlusion, accuracy, resistance to environmental interference, etc. In addition, we try to combine 6D pose estimation and building wind environment assessment to create tangible architectural design approach, we discuss the limitations of the method and point out the direction in which 6d pose estimation is eager to progress in this scenario.
