The Knowing Camera 2: Recognizing And Annotating Places-Of-Interest In Smartphone Photos

Pai Peng,Lidan Shou,Ke Chen,Gang Chen,Sai Wu

DOI: https://doi.org/10.1145/2600428.2609557

2014-01-01

Abstract:This paper presents a project called Knowing Camera for real-time recognizing and annotating places-of-interest(POI) in smartphone photos, with the availability of online geotagged images of such places. We propose a "Spatial+Visual" (S+V) framework which consists of a probabilistic field-of-view model in the spatial phase and sparse coding similarity metric in the visual phase to recognize phone-captured POIs. Moreover, we put forward an offline Collaborative Salient Area (COSTAR) mining algorithm to detect common visual features (called Costars) among the noisy photos geotagged on each POI, thus to clean the geotagged image database. The mining result can be utilized to annotate the region-of-interest on the query image during the online query processing. Besides, this mining procedure further improves the efficiency and accuracy of the S+V framework. Our experiments in the real-world and Oxford 5K datasets show promising recognition and annotation performances of the proposed approach, and that the proposed COSTAR mining technique outperforms state-of-the-art approach.