UC-Net: Uncertainty Inspired RGB-D Saliency Detection via Conditional Variational Autoencoders
This addresses the problem of improving saliency detection accuracy in computer vision for applications like robotics and image analysis, presenting a novel probabilistic approach rather than an incremental improvement.
The paper tackles RGB-D saliency detection by modeling human annotation uncertainty using conditional variational autoencoders, generating multiple saliency maps to produce an accurate consensus, achieving state-of-the-art results on six benchmark datasets against 18 competing algorithms.
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.