Quantifying legibility of indoor spaces using Deep Convolutional Neural Networks: Case studies in train stations

摘要

Legibility is the extent to which space can be easily recognized. Evaluating legibility is particularly desirable in indoor spaces, since it has a large impact on human behavior and the efficiency of space utilization. However, indoor space legibility has only been studied through survey and trivial simulations and lacks scalable quantitative measurement. We utilized a Deep Convolutional Neural Network (DCNN), which is structurally similar to a human perception system, to model legibility in indoor spaces. To implement the modelling of legibility for any indoor space, we designed an end-to-end processing pipeline from indoor data retrieving to model training to spatial legibility analysis. Although the model performed very well (98% accuracy) overall, there are still discrepancies in model's recognizing confidence among different spaces, reflecting legibility differences. To prove the validity of the pipeline, we deployed a survey on Amazon Mechanical Turk, collecting 4015 samples. Meanwhile, we also conducted an identical survey, collecting 570 samples, on occupants in the station. The human samples showed a similar behavior pattern and mechanism as the DCNN models. Further, we used model results to visually explain legibility differences resulting from architectural program, building age, building style, as well as identify visual clusterings of spaces.