Free-Space Detection with Self-Supervised and Online Trained Fully Convolutional Networks
This work addresses the problem of reducing manual annotation effort for free-space detection in autonomous driving, though it is incremental as it builds on existing FCN and self-supervised methods.
The paper tackles free-space detection for vision-based driver assist systems by proposing a Fully Convolutional Network (FCN) trained in a self-supervised manner using stereo-vision disparity to generate labels, reducing the need for manual annotation. It achieves similar results to manual training and shows a 5% performance boost with online training compared to offline training on metrics like Fmax and AP.
Recently, vision-based Advanced Driver Assist Systems have gained broad interest. In this work, we investigate free-space detection, for which we propose to employ a Fully Convolutional Network (FCN). We show that this FCN can be trained in a self-supervised manner and achieve similar results compared to training on manually annotated data, thereby reducing the need for large manually annotated training sets. To this end, our self-supervised training relies on a stereo-vision disparity system, to automatically generate (weak) training labels for the color-based FCN. Additionally, our self-supervised training facilitates online training of the FCN instead of offline. Consequently, given that the applied FCN is relatively small, the free-space analysis becomes highly adaptive to any traffic scene that the vehicle encounters. We have validated our algorithm using publicly available data and on a new challenging benchmark dataset that is released with this paper. Experiments show that the online training boosts performance with 5% when compared to offline training, both for Fmax and AP.