Driven by Convolutional Neural Networks, object detection and semantic segmentation have gained significant improvements. However, existing methods on the basis of a full top-down module have limited robustness in handling those two tasks simultaneously. To this end, we present a joint multi-task framework, termed IvaNet. Different from existing methods, our IvaNet backwards abstract semantic information from higher layers to augment lower layers using local top-down modules. The comparisons against some counterparts on the PASCAL VOC and MS COCO datasets demonstrate the functionality of IvaNet.
@article{arxiv.1903.07360,
title = {IvaNet: Learning to jointly detect and segment objets with the help of Local Top-Down Modules},
author = {Shihua Huang and Lu Wang},
journal= {arXiv preprint arXiv:1903.07360},
year = {2019}
}