# Lightweight RGB-D Salient Object Detection from a Speed-Accuracy Tradeoff Perspective

arXiv: https://arxiv.org/abs/2505.04758 · HTML: https://mlsanity.com/paper/2505.04758/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 34 of its 69 references that are in the map.
- First posted on arXiv 7 May 2025.

## Builds on (most cited first, 30 of 34)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [Squeeze-and-Excitation Networks](https://mlsanity.com/paper/1709.01507/) (2017)
- [MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications](https://mlsanity.com/paper/1704.04861/) (2017)
- [A ConvNet for the 2020s](https://mlsanity.com/paper/2201.03545/) (2022)
- [DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs](https://mlsanity.com/paper/1606.00915/) (2016)
- [CBAM: Convolutional Block Attention Module](https://mlsanity.com/paper/1807.06521/) (2018)
- [Non-local Neural Networks](https://mlsanity.com/paper/1711.07971/) (2017)
- [Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data](https://mlsanity.com/paper/2401.10891/) (2024)
- [Tokens-to-Token ViT: Training Vision Transformers from Scratch on ImageNet](https://mlsanity.com/paper/2101.11986/) (2021)
- [Structure-measure: A New Way to Evaluate Foreground Maps](https://mlsanity.com/paper/1708.00786/) (2017)
- [PraNet: Parallel Reverse Attention Network for Polyp Segmentation](https://mlsanity.com/paper/2006.11392/) (2020)
- [Enhanced-alignment Measure for Binary Foreground Map Evaluation](https://mlsanity.com/paper/1805.10421/) (2018)
- [Dilated Residual Networks](https://mlsanity.com/paper/1705.09914/) (2017)
- [A Simple Pooling-Based Design for Real-Time Salient Object Detection](https://mlsanity.com/paper/1904.09569/) (2019)
- [ACNet: Strengthening the Kernel Skeletons for Powerful CNN via Asymmetric Convolution Blocks](https://mlsanity.com/paper/1908.03930/) (2019)
- [Rethinking RGB-D Salient Object Detection: Models, Data Sets, and Large-Scale Benchmarks](https://mlsanity.com/paper/1907.06781/) (2019)
- [Non-Salient Region Object Mining for Weakly Supervised Semantic Segmentation](https://mlsanity.com/paper/2103.14581/) (2021)
- [Siamese Network for RGB-D Salient Object Detection and Beyond](https://mlsanity.com/paper/2008.12134/) (2020)
- [SwinNet: Swin Transformer drives edge-aware RGB-D and RGB-T salient object detection](https://mlsanity.com/paper/2204.05585/) (2022)
- [Deep RGB-D Saliency Detection with Depth-Sensitive Attention and Automatic Multi-Modal Fusion](https://mlsanity.com/paper/2103.11832/) (2021)
- [CAVER: Cross-Modal View-Mixed Transformer for Bi-Modal Salient Object Detection](https://mlsanity.com/paper/2112.02363/) (2021)
- [TriTransNet: RGB-D Salient Object Detection with a Triplet Transformer Embedding Network](https://mlsanity.com/paper/2108.03990/) (2021)
- [Bilateral Attention Network for RGB-D Salient Object Detection](https://mlsanity.com/paper/2004.14582/) (2020)
- [Depth Quality-Inspired Feature Manipulation for Efficient RGB-D Salient Object Detection](https://mlsanity.com/paper/2107.01779/) (2021)
- [HiDAnet: RGB-D Salient Object Detection via Hierarchical Depth Awareness](https://mlsanity.com/paper/2301.07405/) (2023)
- [Non-rigid Object Tracking via Deep Multi-scale Spatial-temporal Discriminative Saliency Maps](https://mlsanity.com/paper/1802.07957/) (2018)
- [MobileSal: Extremely Efficient RGB-D Salient Object Detection](https://mlsanity.com/paper/2012.13095/) (2020)
- [Point-aware Interaction and CNN-induced Refinement Network for RGB-D Salient Object Detection](https://mlsanity.com/paper/2308.08930/) (2023)

## Built on by (most cited first, 6 of 6)

- [Controllable-LPMoE: Adapting to Challenging Object Segmentation via Dynamic Local Priors from Mixture-of-Experts](https://mlsanity.com/paper/2510.21114/) (2025)
- [Dual-Perspective United Transformer for Object Segmentation in Optical Remote Sensing Images](https://mlsanity.com/paper/2506.21866/) (2025)
- [HVPNet: A Bio-Inspired Network for General Salient and Camouflaged Object Detection](https://mlsanity.com/paper/2606.31496/) (2026)
- [Propagating Sparse Depth via Depth Foundation Model for Out-of-Distribution Depth Completion](https://mlsanity.com/paper/2508.04984/) (2025)
- [Learning Spatially Adaptive Structural Coordination for Underwater Salient Object Detection](https://mlsanity.com/paper/2605.15535/) (2026)
- [VCP-DCN: Beyond Visual Concealed Property via Depth Collaborative Network for Camouflaged Object Detection](https://mlsanity.com/paper/2607.27843/) (2026)

## Abstract (arXiv)

Current RGB-D methods usually leverage large-scale backbones to improve accuracy but sacrifice efficiency. Meanwhile, several existing lightweight methods are difficult to achieve high-precision performance. To balance the efficiency and performance, we propose a Speed-Accuracy Tradeoff Network (SATNet) for Lightweight RGB-D SOD from three fundamental perspectives: depth quality, modality fusion, and feature representation. Concerning depth quality, we introduce the Depth Anything Model to generate high-quality depth maps,which effectively alleviates the multi-modal gaps in the current datasets. For modality fusion, we propose a Decoupled Attention Module (DAM) to explore the consistency within and between modalities. Here, the multi-modal features are decoupled into dual-view feature vectors to project discriminable information of feature maps. For feature representation, we develop a Dual Information Representation Module (DIRM) with a bi-directional inverted framework to enlarge the limited feature space generated by the lightweight backbones. DIRM models texture features and saliency features to enrich feature space, and employ two-way prediction heads to optimal its parameters through a bi-directional backpropagation. Finally, we design a Dual Feature Aggregation Module (DFAM) in the decoder to aggregate texture and saliency features. Extensive experiments on five public RGB-D SOD datasets indicate that the proposed SATNet excels state-of-the-art (SOTA) CNN-based heavyweight models and achieves a lightweight framework with 5.2 M parameters and 415 FPS.