RGB D Salient Object Detection Based on Discriminative Cross modal Transfer Learning

In this work, we propose to utilize Convolutional Neural Networks to boost the performance of depth-induced salient object detection . We use CNN-based cross-modal transfer problem to bridge the gap between the “data-hungry” nature of CNNs and the unavailability of sufficient labeled training data in depth modality . We exploit the depth-specific information by pre-training a modality classification network that encourages modal-specific representations during the optimizing course . These two modules are pre-trained independently and then stitched to initialize and optimize the eventual deep-induced saliency detection model. Experiments demonstrate the effectiveness of the proposed novel pre-train strategy as well as the proposed approach over other state-of-the-art methods. The proposed approach is described as a novel pre trainings strategy and the significant and consistent improvements of

Links: PDF - Abstract

Code :

None

Keywords : pre - depth - modal - detection - proposed -

Leave a Reply

Your email address will not be published. Required fields are marked *