mlsanityData to 14 Oct 2026

Towards Size-invariant Salient Object Detection: A Generic Evaluation and Optimization Approach

arXiv 2509.15573 ↗First posted 19 Sep 2025cs.CV cs.AI cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

This paper investigates a fundamental yet underexplored issue in Salient Object Detection (SOD): the size-invariant property for evaluation protocols, particularly in scenarios when multiple salient objects of significantly different sizes appear within a single image. We first present a novel perspective to expose the inherent size sensitivity of existing widely used SOD metrics. Through careful theoretical derivations, we show that the evaluation outcome of an image under current SOD metrics can be essentially decomposed into a sum of several separable terms, with the contribution of each term being directly proportional to its corresponding region size. Consequently, the prediction errors would be dominated by the larger regions, while smaller yet potentially more semantically important objects are often overlooked, leading to biased performance assessments and practical degradation. To address this challenge, a generic Size-Invariant Evaluation (SIEva) framework is proposed. The core idea is to evaluate each separable component individually and then aggregate the results, thereby effectively mitigating the impact of size imbalance across objects. Building upon this, we further develop a dedicated optimization framework (SIOpt), which adheres to the size-invariant principle and significantly enhances the detection of salient objects across a broad range of sizes. Notably, SIOpt is model-agnostic and can be seamlessly integrated with a wide range of SOD backbones. Theoretically, we also present generalization analysis of SOD methods and provide evidence supporting the validity of our new evaluation protocols. Finally, comprehensive experiments speak to the efficacy of our proposed approach. The code is available at https://github.com/Ferry-Li/SI-SOD.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 68 of its 138 references that are in the map. First posted on arXiv 19 Sep 2025.

Builds on

68 of 138 references.

References · 68

  1. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  2. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,639
  3. 2023Segment Anythingcited by 8,234
  4. 2015You Only Look Once: Unified, Real-Time Object Detectioncited by 4,852
  5. 2016V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentationcited by 2,292
  6. 2013Rich feature hierarchies for accurate object detection and semantic segmentationcited by 2,162
  7. 2017Structure-measure: A New Way to Evaluate Foreground Mapscited by 480
  8. 2018Enhanced-alignment Measure for Binary Foreground Map Evaluationcited by 370
  9. 2020Causal Intervention for Weakly-Supervised Semantic Segmentationcited by 229
  10. 2019A Simple Pooling-Based Design for Real-Time Salient Object Detectioncited by 212
  11. 2016UnitBox: An Advanced Object Detection Networkcited by 205
  12. 2021Object Detection in Aerial Images: A Large-Scale Benchmark and Challengescited by 204
38 more
  1. 2015Visual Saliency Based on Multiscale Deep Featurescited by 185
  2. 2016What do different evaluation metrics tell us about saliency models?cited by 185
  3. 2021Visual Saliency Transformercited by 175
  4. 2019Salient Object Detection in the Deep Learning Era: An In-Depth Surveycited by 164
  5. 2016A vector-contraction inequality for Rademacher complexitiescited by 142
  6. 2020Few-Shot Object Detection and Viewpoint Estimation for Objects in the Wildcited by 142
  7. 2021Salient Object Detection via Integrity Learningcited by 133
  8. 2015Salient Object Detection: A Benchmarkcited by 131
  9. 2020Suppress and Balance: A Simple Gated Network for Salient Object Detectioncited by 126
  10. 2020Label Decoupling Framework for Salient Object Detectioncited by 111
  11. 2022AUC Maximization in the Era of Big Data and AI: A Surveycited by 100
  12. 2014Salient Object Detection: A Surveycited by 98
  13. 2017Amulet: Aggregating Multi-level Convolutional Features for Salient Object Detectioncited by 93
  14. 2020Large-scale Robust Deep AUC Maximization: A New Surrogate Loss and Empirical Studies on Medical Image Classificationcited by 86
  15. 2021Sample Selection with Uncertainty of Losses for Learning with Noisy Labelscited by 79
  16. 2020RGBT Salient Object Detection: A Large-scale Dataset and Benchmarkcited by 69
  17. 2020EDN: Salient Object Detection via Extremely-Downsampled Networkcited by 64
  18. 2022SwinNet: Swin Transformer drives edge-aware RGB-D and RGB-T salient object detectioncited by 59
  19. 2019RGB-T Image Saliency Detection via Collaborative Graph Learningcited by 58
  20. 2021Learning Generative Vision Transformer with Energy-Based Latent Space for Saliency Predictioncited by 57
  21. 2016Deep Contrast Learning for Salient Object Detectioncited by 52
  22. 2020Accurate RGB-D Salient Object Detection via Collaborative Learningcited by 52
  23. 2017A Unified RGB-T Saliency Detection Benchmark: Dataset, Baselines, Analysis and A Novel Approachcited by 51
  24. 2021TriTransNet: RGB-D Salient Object Detection with a Triplet Transformer Embedding Networkcited by 45
  25. 2017Learning Uncertain Convolutional Features for Accurate Saliency Detectioncited by 44
  26. 2022Does Thermal Really Always Matter for RGB-T Salient Object Detection?cited by 37
  27. 2020A General Framework for Consistent Structured Prediction with Implicit Loss Embeddingscited by 35
  28. 2021Learning with Multiclass AUC: Theory and Algorithmscited by 35
  29. 2018PiCANet: Pixel-wise Contextual Attention Learning for Accurate Saliency Detectioncited by 28
  30. 2023VST++: Efficient and Stronger Visual Saliency Transformercited by 28
  31. 2015DeepSaliency: Multi-Task Deep Neural Network Model for Salient Object Detectioncited by 24
  32. 2024A Simple yet Effective Network based on Vision Transformer for Camouflaged Object and Salient Object Detectioncited by 23
  33. 2024Learning Adaptive Fusion Bank for Multi-modal Salient Object Detectioncited by 19
  34. 2020Regularized Densely-connected Pyramid Network for Salient Instance Segmentationcited by 18
  35. 2024AUCSeg: AUC-oriented Pixel-level Long-tail Semantic Segmentationcited by 18
  36. 2023Revisiting Evaluation Metrics for Semantic Segmentation: Optimization and Evaluation of Fine-grained Intersection over Unioncited by 17
  37. 2022Minimax AUC Fairness: Efficient Algorithm with Provable Convergencecited by 16
  38. 2022Optimizing Two-way Partial AUC with an End-to-end Frameworkcited by 15

50 most cited of 68 All 68 (JSON)

Built on by

Not cited yet.