Deep Hypersphere Feature Regularization for Weakly Supervised RGB-D Salient Object Detection

doi:10.1109/TIP.2023.3318953

Deep Hypersphere Feature Regularization for Weakly Supervised RGB-D Salient Object Detection

2023-01-01

4

PRE

AI

DOI：10.1109/TIP.2023.3318953

Original

Original request for help

Save

Abstract

En 中文

We propose a weakly supervised approach for salient object detection from multi-modal RGB-D data. Our approach only relies on labels from scribbles, which are much easier to annotate, compared with dense labels used in conventional fully supervised setting. In contrast to existing methods that employ supervision signals on the output space, our design regularizes the intermediate latent space to enhance discrimination between salient and non-salient objects. We further introduce a contour detection branch to implicitly constrain the semantic boundaries and achieve precise edges of detected salient objects. To enhance the long-range dependencies among local features, we introduce a Cross-Padding Attention Block (CPAB). Extensive experiments on seven benchmark datasets demonstrate that our method not only outperforms existing weakly supervised methods, but is also on par with several fully-supervised state-of-the-art models. Code is available at https://github.com/leolyj/DHFR-SOD.

Keywords:

Feature extraction

Semantics

Transformers

Object detection

Decoding

Annotations

Image edge detection

Salient object detection

weakly supervised learning

Deep Hypersphere Feature Regularization

Von Mises Fisher

Deep Hypersphere Feature Regularization for Weakly Supervised RGB-D Salient Object Detection

Abstract

Journal

Organization