# SupScene: Scene-Structured Overlap Supervision for Image Retrieval in Unconstrained SfM

arXiv: https://arxiv.org/abs/2601.11930 · HTML: https://mlsanity.com/paper/2601.11930/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 38 references that are in the map.
- First posted on arXiv 17 Jan 2026.

## Builds on (most cited first, 18 of 18)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [NetVLAD: CNN architecture for weakly supervised place recognition](https://mlsanity.com/paper/1511.07247/) (2015)
- [Fine-tuning CNN Image Retrieval with No Human Annotation](https://mlsanity.com/paper/1711.02512/) (2017)
- [BlendedMVS: A Large-scale Dataset for Generalized Multi-view Stereo Networks](https://mlsanity.com/paper/1911.10127/) (2019)
- [CNN Features off-the-shelf: an Astounding Baseline for Recognition](https://mlsanity.com/paper/1403.6382/) (2014)
- [Learning with Average Precision: Training Image Retrieval with a Listwise Loss](https://mlsanity.com/paper/1906.07589/) (2019)
- [AnyLoc: Towards Universal Visual Place Recognition](https://mlsanity.com/paper/2308.00688/) (2023)
- [Particular object retrieval with integral max-pooling of CNN activations](https://mlsanity.com/paper/1511.05879/) (2015)
- [Deep Image Retrieval: Learning global representations for image search](https://mlsanity.com/paper/1604.01325/) (2016)
- [Aggregating Deep Convolutional Features for Image Retrieval](https://mlsanity.com/paper/1510.07493/) (2015)
- [CNN Image Retrieval Learns from BoW: Unsupervised Fine-Tuning with Hard Examples](https://mlsanity.com/paper/1604.02426/) (2016)
- [DOLG: Single-Stage Image Retrieval with Deep Orthogonal Fusion of Local and Global Features](https://mlsanity.com/paper/2108.02927/) (2021)
- [Correlation Verification for Image Retrieval](https://mlsanity.com/paper/2204.01458/) (2022)
- [Cross-dimensional Weighting for Aggregated Deep Convolutional Features](https://mlsanity.com/paper/1512.04065/) (2015)
- [ILIAS: Instance-Level Image retrieval At Scale](https://mlsanity.com/paper/2502.11748/) (2025)
- [Matchable Image Retrieval by Learning from Surface Reconstruction](https://mlsanity.com/paper/1811.10343/) (2018)
- [Image Retrieval for Structure-from-Motion via Graph Convolutional Network](https://mlsanity.com/paper/2009.08049/) (2020)

## Built on by (most cited first, 1 of 1)

- [On-the-Fly3R: Towards Robust Online 3D Reconstruction with Feed-Forward 3R Models for Large-Scale UAV Scenarios](https://mlsanity.com/paper/2609.00923/) (2026)

## Abstract (arXiv)

Image retrieval is a critical step for reducing the quadratic cost of image matching in unconstrained Structure-from-Motion (SfM). Unlike generic image retrieval, however, the relevant goal of SfM is to identify geometrically matchable image pairs rather than merely semantically similar images. Prevailing methods are largely trained under anchor-centric tuple guidance, which organizes the training around isolated tuples and under-utilizes the dense, graded overlap structure naturally established within a SfM scene. In this work, we present SupScene, a scene-structured training framework that samples connected local subgraphs from SfM overlap graphs and jointly supervises all valid within-subgraph pairwise relations. To explicitly align the trained descriptor with geometric co-visibility, we further introduce an overlap-ordered objective that combines multi-similarity optimization with a continuous relative-overlap ranking term. In addition, the proposed framework is instantiated with a lightweight Structural Context Probe Pooling (SCPP) head that aggregates complementary structural responses into a compact global descriptor. Extensive experimental results on multiple benchmarks demonstrate that our method can significantly improve overall retrieval performance and enhance the completeness of downstream SfM reconstructions. Code and models are available at https://github.com/Suxilan/SupScene.