Sparse weakly supervised models for object localization in road environment

Abstract

We propose a novel weakly supervised localization method based on Fisher-embedding of low-level features (CNN, SIFT), and model sparsity at the component level. Fisher-embedding provides an interesting alternative to raw lowlevel features, since it allows fast and accurate scoring of image subwindows with a model trained on entire images. Model sparsity reduces overfitting and enables fast evaluation. We also propose two new techniques for improving performance when our method is combined with nonlinear normalizations of the aggregated Fisher representation of the image. These techniques are i) intra-component metric normalization and ii) first-order approximation to the score of a normalized image representation. We evaluate our weakly supervised localization method on real traffic scenes acquired from driver's perspective. The method dramatically improves the localization AP over the dense nonnormalized Fisher vector baseline (16 percentage points for zebra crossings, 21 percentage points for traffic signs) and leads to a huge gain in execution speed (91× for zebra crossings, 74× for traffic signs).

Assessment of bias in a crowdsourcing accessibility system using serious games

“…A different approach is to apply image-processing algorithms to automatically extract geo-3.6 graphical accessibility data, e.g., mapping pedestrian zebra crossings (Ahmetovic et al, 2017;Berriel et al, 2017;Coughlan and Shen, 2013;Guy and Truong, 2012;Haider et al, 2019;Koester et al, 2016;Zadrija et al, 2018). These systems are capable of extracting candidate crossings from satellite or street-level images, and their overall accuracy can be increased by combining both kinds of images and using crowdsourcing to further validate acquired data (Ahmetovic et al, 2017).…”