Spatial Pattern Templates for Recognition of Objects with Regular Structure

Abstract

We propose a method for semantic parsing of images with regular structure. The structured objects are modeled in a densely connected CRF. The paper describes how to embody specific spatial relations in a representation called Spatial Pattern Templates (SPT), which allows us to capture regularity constraints of alignment and equal spacing in pairwise and ternary potentials. Assuming the input image is pre-segmented to salient regions the SPT describe which segments could interact in the structured graphical model. The model parameters are learnt to describe the formal language of semantic labelings. Given an input image, a consistent labeling over its segments linked in the CRF is recognized as a word from this language. The CRF framework allows us to apply efficient algorithms for both recognition and learning. We demonstrate the approach on the problem of facade image parsing and show that results comparable with state of the art methods are achieved without introducing additional manually designed detectors for specific terminal objects.

Detecting repetitive elements with accurate locations and shapes from urban façade

“…Our test databases originate from different sources: the ECP-Monge database [20], Zübu database [21] and our selected façade database. We manually get rid of some pictures in which facades are too small or too much occluded.…”

“…Fig. 5 The comparison results in terms of Precision, Recall rates and F1-score by using databases [20,21] and our database.…”

Large Scale Asset Extraction for Urban Images

“…Datasets. We use the following datasets for evaluation: GRAZ50 [20], ECP [24], eTrims [12], CMP [26] and two new datasets for London and Strasbourg that we manually compiled comprising 800 images taken in the wild. The total dataset size is 1392 images.…”

A Convolutional Network for Semantic Facade Segmentation and Interpretation

“…The detection of regular window patterns has been improved by (Tyleček andŠára, 2012) by guiding it with a stochastic grammar with pair-wise attribute constraints. In (Tyleček andŠára, 2013), additionally a data-dependent topology of spatial templates has been introduced. (Simon et al, 2011) segment facades into windows and doors as well as a couple of other objects.…”