User-Assisted Inverse Procedural Facade Modeling and Compressed Image Rendering

We take advantage of human intuition by encoding facades into a procedural representation. Our user-assisted inverse procedural modeling approach allows users to exploit repetitions and symmetries of facades to create a split grammar representation of the input. Terminal symbols correspond to repeating elements such as windows, window panes, and doors and their distributions are encoded as the production rules. Our participants achieved a compression factor that averaged 57% (min=12%, max=99%) while taking on average 7 minutes (min=1, max=25) to compress an image. The compressed facades do not suffer from occlusion problems present in the input, such as trees or cars. Our second contribution is a novel rendering algorithm that directly displays the compressed facades in their procedural form by interpreting the procedural rules during texture lookup. This algorithm provides considerable memory savings while achieving comparable rendering performance. Terminal Symbols Input image Procedural Representation a) 2.43 MB b) c) 1.27 MB CF=48.7% d) e) 366.5 kB CF=85.3% Fig. 1. The user encoded a facade (a) procedurally with attention paid to conveying the varying details (b) resulting in 44 procedural rules, 49% compression, and 10 terminal symbols (c). The same facade was encoded again with less precision (d) resulting in only 30 procedural rules, 85% compression, and 9 terminal symbols (e). The user removed the occluding trees and each encoding took under two minutes. The original image was rendered in 1.32 ms and the direct compressed facade rendering was 1.73 ms for (b) and 1.54 ms for (d). 2 Huilong Zhuo, Shengchuan Zhou, Bedrich Benes, and David Whittinghill