UC-Net: Uncertainty Inspired RGB-D Saliency Detection via Conditional Variational Autoencoders

Abstract

In this paper, we propose the first framework (UCNet) to employ uncertainty for RGB-D saliency detection by learning from the data labeling process. Existing RGB-D saliency detection methods treat the saliency detection task as a point estimation problem, and produce a single saliency map following a deterministic learning pipeline. Inspired by the saliency data labeling process, we propose probabilistic RGB-D saliency detection network via conditional variational autoencoders to model human annotation uncertainty and generate multiple saliency maps for each input image by sampling in the latent space. With the proposed saliency consensus process, we are able to generate an accurate saliency map based on these multiple predictions. Quantitative and qualitative evaluations on six challenging benchmark datasets against 18 competing algorithms demonstrate the effectiveness of our approach in learning the distribution of saliency maps, leading to a new state-of-the-art in RGB-D saliency detection.

arXiv commentAccepted by IEEE CVPR 2020 (ORAL). Code: https://github.com/JingZhang617/UCNet

References in corpus (6)

• Auto-Encoding Variational Bayes
• Rethinking RGB-D Salient Object Detection: Models, Data Sets, and Large-Scale Benchmarks
• PixelVAE: A Latent Variable Model for Natural Images
• Deep Unsupervised Saliency Detection: A Multiple Noisy Labeling Perspective
• Contrastive Variational Autoencoder Enhances Salient Features
• Learning Variations in Human Motion via Mix-and-Match Perturbation

Cited by in corpus (6)

• Contrast-weighted Dictionary Learning Based Saliency Detection for Remote Sensing Images
• JL-DCF: Joint Learning and Densely-Cooperative Fusion Framework for RGB-D Salient Object Detection
• Siamese Network for RGB-D Salient Object Detection and Beyond
• Improving Camouflaged Object Detection with the Uncertainty of Pseudo-edge Labels
• Depth Quality-Inspired Feature Manipulation for Efficient RGB-D Salient Object Detection
• Densely Deformable Efficient Salient Object Detection Network