Rethinking Lightweight Salient Object Detection via Network Depth-Width Tradeoff

TL;DR

This paper introduces a lightweight salient object detection framework that balances efficiency and accuracy by decoupling the U-shape structure into three branches and optimizing network depth and width for different application needs.

Contribution

The authors propose a novel trilateral decoder framework and a scale-adaptive pooling module, enabling effective lightweight models without additional parameters, and explore the depth-width tradeoff for SOD.

Findings

Achieves high FPS on resource-constrained devices

Outperforms existing methods on five benchmarks

Offers multiple model variants for different application scenarios

Abstract

Existing salient object detection methods often adopt deeper and wider networks for better performance, resulting in heavy computational burden and slow inference speed. This inspires us to rethink saliency detection to achieve a favorable balance between efficiency and accuracy. To this end, we design a lightweight framework while maintaining satisfying competitive accuracy. Specifically, we propose a novel trilateral decoder framework by decoupling the U-shape structure into three complementary branches, which are devised to confront the dilution of semantic context, loss of spatial structure and absence of boundary detail, respectively. Along with the fusion of three branches, the coarse segmentation results are gradually refined in structure details and boundary quality. Without adding additional learnable parameters, we further propose Scale-Adaptive Pooling Module to obtain multi-scale receptive filed. In particular, on the premise of inheriting this framework, we rethink the relationship among accuracy, parameters and speed via network depth-width tradeoff. With these insightful considerations, we comprehensively design shallower and narrower models to explore the maximum potential of lightweight SOD. Our models are purposed for different application environments: 1) a tiny version CTD-S (1.7M, 125FPS) for resource constrained devices, 2) a fast version CTD-M (12.6M, 158FPS) for speed-demanding scenarios, 3) a standard version CTD-L (26.5M, 84FPS) for high-performance platforms. Extensive experiments validate the superiority of our method, which achieves better efficiency-accuracy balance across five benchmarks.