# Exploring the Limits of End-to-End Feature-Affinity Propagation for Single-Point Supervised Infrared Small Target Detection

arXiv: https://arxiv.org/abs/2605.00722 · HTML: https://mlsanity.com/paper/2605.00722/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 13 references that are in the map.
- First posted on arXiv 1 May 2026.

## Builds on (most cited first, 9 of 9)

- [Learning Pixel-level Semantic Affinity with Image-level Supervision for Weakly Supervised Semantic Segmentation](https://mlsanity.com/paper/1803.10464/) (2018)
- [Semi-Supervised Semantic Segmentation Using Unreliable Pseudo-Labels](https://mlsanity.com/paper/2203.03884/) (2022)
- [Mapping Degeneration Meets Label Evolution: Learning Infrared Small Target Detection with Single Point Supervision](https://mlsanity.com/paper/2304.01484/) (2023)
- [Monte Carlo Linear Clustering with Single-Point Supervision is Enough for Infrared Small Target Detection](https://mlsanity.com/paper/2304.04442/) (2023)
- [From Easy to Hard: Progressive Active Learning Framework for Infrared Small Target Detection with Single Point Supervision](https://mlsanity.com/paper/2412.11154/) (2024)
- [Label-efficient Segmentation via Affinity Propagation](https://mlsanity.com/paper/2310.10533/) (2023)
- [When Confidence Fails: Revisiting Pseudo-Label Selection in Semi-supervised Semantic Segmentation](https://mlsanity.com/paper/2509.16704/) (2025)
- [Point-to-Mask: From Arbitrary Point Annotations to Mask-Level Infrared Small Target Detection](https://mlsanity.com/paper/2603.16257/) (2026)
- [Rethinking IRSTD: Single-Point Supervision Guided Encoder-only Framework is Enough for Infrared Small Target Detection](https://mlsanity.com/paper/2604.05363/) (2026)

## Abstract (arXiv)

Single-point supervised infrared small target detection (IRSTD) drastically reduces dense annotation costs. Current state-of-the-art (SOTA) methods achieve high precision by recovering mask supervision through explicit, offline pseudo-label construction, such as multi-stage active learning and physics-driven mask generation. In this paper, we study a minimalist alternative: generating point-to-mask supervision online through in-batch, point-anchored feature-affinity propagation. We instantiate this paradigm as GSACP, an end-to-end testbed that directly supervises the detector using hard-margin feature affinity gated by local image priors, entirely eliminating external label-evolution loops. This compact design, however, exposes an optimization bottleneck. Because the affinity target is generated from the same feature representation being optimized, training forms a self-referential loop. We theoretically formalize this as Self-Referential Propagation Drift, a representation-supervision entanglement that can sharpen true boundaries or distort the feature space to satisfy its own targets. To systematically isolate these failure modes, we apply a protocolized single-variable ablation procedure spanning local EMA teacher decoupling, hard-background contrastive separation, and adaptive support geometry. On the SIRST3 dataset, GSACP-Final establishes a new ultra-low false-alarm operating regime, achieving a highly competitive 0.6674 mIoU while demonstrating a 38% relative reduction in false-positive artifacts (Fa) compared with PAL. By systematically deconstructing the end-to-end paradigm, we map its performance boundaries and show that in-batch feature propagation provides a compact alternative for deployment scenarios where false-alarm suppression is paramount.