Dynamic Knowledge Distillation With Noise Elimination for RGB-D Salient Object Detection

TL;DR

This paper introduces a dynamic knowledge distillation approach with noise elimination for RGB-D salient object detection, reducing computational costs while maintaining high accuracy and speed.

Contribution

It proposes a lightweight, dynamic distillation method that considers teacher and student performance and introduces noise elimination to handle low-quality depth data.

Findings

Achieves competitive accuracy on five datasets.

Runs at 136 FPS, faster than previous methods.

Effectively mitigates noise from low-quality depth maps.

Abstract

RGB-D salient object detection (SOD) demonstrates its superiority on detecting in complex environments due to the additional depth information introduced in the data. Inevitably, an independent stream is introduced to extract features from depth images, leading to extra computation and parameters. This methodology sacrifices the model size to improve the detection accuracy which may impede the practical application of SOD problems. To tackle this dilemma, we propose a dynamic distillation method along with a lightweight structure, which significantly reduces the computational burden while maintaining validity. This method considers the factors of both teacher and student performance within the training stage and dynamically assigns the distillation weight instead of applying a fixed weight on the student model. We also investigate the issue of RGB-D early fusion strategy in distillation and propose a simple noise elimination method to mitigate the impact of distorted training data caused by low quality depth maps. Extensive experiments are conducted on five public datasets to demonstrate that our method can achieve competitive performance with a fast inference speed (136FPS) compared to 10 prior methods.