# Split-Fuse-Transport: Annotation-Free Saliency via Dual Clustering and Optimal Transport Alignment

arXiv: https://arxiv.org/abs/2510.17484 · HTML: https://mlsanity.com/paper/2510.17484/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 28 of its 43 references that are in the map.
- First posted on arXiv 20 Oct 2025.

## Builds on (most cited first, 28 of 28)

- [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)
- [DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs](https://mlsanity.com/paper/1606.00915/) (2016)
- [Learning Deep Features for Discriminative Localization](https://mlsanity.com/paper/1512.04150/) (2015)
- [Per-Pixel Classification is Not All You Need for Semantic Segmentation](https://mlsanity.com/paper/2107.06278/) (2021)
- [U2-Net: Going Deeper with Nested U-Structure for Salient Object Detection](https://mlsanity.com/paper/2005.09007/) (2020)
- [EGNet:Edge Guidance Network for Salient Object Detection](https://mlsanity.com/paper/1908.08297/) (2019)
- [Localizing Objects with Self-Supervised Transformers and no Labels](https://mlsanity.com/paper/2109.14279/) (2021)
- [Simple Does It: Weakly Supervised Instance and Semantic Segmentation](https://mlsanity.com/paper/1603.07485/) (2016)
- [Deeply supervised salient object detection with short connections](https://mlsanity.com/paper/1611.04849/) (2016)
- [Visual Saliency Transformer](https://mlsanity.com/paper/2104.12099/) (2021)
- [Self-Supervised Transformers for Unsupervised Object Discovery using Normalized Cut](https://mlsanity.com/paper/2202.11539/) (2022)
- [Suppress and Balance: A Simple Gated Network for Salient Object Detection](https://mlsanity.com/paper/2007.08074/) (2020)
- [TokenCut: Segmenting Objects in Images and Videos with Self-supervised Transformer and Normalized Cut](https://mlsanity.com/paper/2209.00383/) (2022)
- [Weakly-Supervised Salient Object Detection via Scribble Annotations](https://mlsanity.com/paper/2003.07685/) (2020)
- [EDN: Salient Object Detection via Extremely-Downsampled Network](https://mlsanity.com/paper/2012.13093/) (2020)
- [Unsupervised Salient Object Detection with Spectral Cluster Voting](https://mlsanity.com/paper/2203.12614/) (2022)
- [Learning Uncertain Convolutional Features for Accurate Saliency Detection](https://mlsanity.com/paper/1708.02031/) (2017)
- [Multi-source weak supervision for saliency detection](https://mlsanity.com/paper/1904.00566/) (2019)
- [Weakly Supervised Salient Object Detection Using Image Labels](https://mlsanity.com/paper/1803.06503/) (2018)
- [Learning Noise-Aware Encoder-Decoder from Noisy Labels by Alternating Back-Propagation for Saliency Detection](https://mlsanity.com/paper/2007.12211/) (2020)
- [MFNet: Multi-filter Directive Network for Weakly Supervised Salient Object Detection](https://mlsanity.com/paper/2112.01732/) (2021)
- [SFC: Shared Feature Calibration in Weakly Supervised Semantic Segmentation](https://mlsanity.com/paper/2401.11719/) (2024)
- [Unsupervised Domain Adaptive Salient Object Detection Through Uncertainty-Aware Pseudo-Label Learning](https://mlsanity.com/paper/2202.13170/) (2022)
- [A Mutual Learning Method for Salient Object Detection with intertwined Multi-Supervision--Revised](https://mlsanity.com/paper/2509.21363/) (2025)
- [TDIOT: Target-driven Inference for Deep Video Object Tracking](https://mlsanity.com/paper/2103.11017/) (2021)
- [Locally-Focused Face Representation for Sketch-to-Image Generation Using Noise-Induced Refinement](https://mlsanity.com/paper/2411.19005/) (2024)
- [Gated-Attention Feature-Fusion Based Framework for Poverty Prediction](https://mlsanity.com/paper/2411.19690/) (2024)

## Abstract (arXiv)

Salient object detection (SOD) aims to segment visually prominent regions in images and serves as a foundational task for various computer vision applications. We posit that SOD can now reach near-supervised accuracy without a single pixel-level label, but only when reliable pseudo-masks are available. We revisit the prototype-based line of work and make two key observations. First, boundary pixels and interior pixels obey markedly different geometry; second, the global consistency enforced by optimal transport (OT) is underutilized if prototype quality is weak. To address this, we introduce POTNet, an adaptation of Prototypical Optimal Transport that replaces POT's single k-means step with an entropy-guided dual-clustering head: high-entropy pixels are organized by spectral clustering, low-entropy pixels by k-means, and the two prototype sets are subsequently aligned by OT. This split-fuse-transport design yields sharper, part-aware pseudo-masks in a single forward pass, without handcrafted priors. Those masks supervise a standard MaskFormer-style encoder-decoder, giving rise to AutoSOD, an end-to-end unsupervised SOD pipeline that eliminates SelfMask's offline voting yet improves both accuracy and training efficiency. Extensive experiments on five benchmarks show that AutoSOD outperforms unsupervised methods by up to 26% and weakly supervised methods by up to 36% in F-measure, further narrowing the gap to fully supervised models.