# What Do Scan-Derived Class Prototypes Add? Disentangling Supervision, Prototype Content and Query Protocol in Recognition over Frozen Foundation Features

arXiv: https://arxiv.org/abs/2609.04381 · HTML: https://mlsanity.com/paper/2609.04381/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 30 of its 40 references that are in the map.
- First posted on arXiv 3 Sep 2026.

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

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation](https://mlsanity.com/paper/1612.00593/) (2016)
- [PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space](https://mlsanity.com/paper/1706.02413/) (2017)
- [Prototypical Networks for Few-shot Learning](https://mlsanity.com/paper/1703.05175/) (2017)
- [SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features](https://mlsanity.com/paper/2502.14786/) (2025)
- [Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecture](https://mlsanity.com/paper/2301.08243/) (2023)
- [A Closer Look at Few-shot Classification](https://mlsanity.com/paper/1904.04232/) (2019)
- [Recognition in Terra Incognita](https://mlsanity.com/paper/1807.04975/) (2018)
- [Point-BERT: Pre-training 3D Point Cloud Transformers with Masked Point Modeling](https://mlsanity.com/paper/2111.14819/) (2021)
- [Rethinking Few-Shot Image Classification: a Good Embedding Is All You Need?](https://mlsanity.com/paper/2003.11539/) (2020)
- [FoundationPose: Unified 6D Pose Estimation and Tracking of Novel Objects](https://mlsanity.com/paper/2312.08344/) (2023)
- [Noise or Signal: The Role of Image Backgrounds in Object Recognition](https://mlsanity.com/paper/2006.09994/) (2020)
- [A Metric Learning Reality Check](https://mlsanity.com/paper/2003.08505/) (2020)
- [T-LESS: An RGB-D Dataset for 6D Pose Estimation of Texture-less Objects](https://mlsanity.com/paper/1701.05498/) (2017)
- [Cut, Paste and Learn: Surprisingly Easy Synthesis for Instance Detection](https://mlsanity.com/paper/1708.01642/) (2017)
- [MegaPose: 6D Pose Estimation of Novel Objects via Render & Compare](https://mlsanity.com/paper/2212.06870/) (2022)
- [BOP Challenge 2020 on 6D Object Localization](https://mlsanity.com/paper/2009.07378/) (2020)
- [OpenShape: Scaling Up 3D Shape Representation Towards Open-World Understanding](https://mlsanity.com/paper/2305.10764/) (2023)
- [OnePose: One-Shot Object Pose Estimation without CAD Models](https://mlsanity.com/paper/2205.12257/) (2022)
- [SparseGS: Sparse View Synthesis using 3D Gaussian Splatting](https://mlsanity.com/paper/2312.00206/) (2023)
- [GigaPose: Fast and Robust Novel Object Pose Estimation via One Correspondence](https://mlsanity.com/paper/2311.14155/) (2023)
- [FoundPose: Unseen Object Pose Estimation with Foundation Features](https://mlsanity.com/paper/2311.18809/) (2023)
- [CNOS: A Strong Baseline for CAD-based Novel Object Segmentation](https://mlsanity.com/paper/2307.11067/) (2023)
- [Hyperspherical Prototype Networks](https://mlsanity.com/paper/1901.10514/) (2019)
- [Fix your classifier: the marginal value of training the last weight layer](https://mlsanity.com/paper/1801.04540/) (2018)
- [Shape-biased Texture Agnostic Representations for Improved Textureless and Metallic Object Detection and 6D Pose Estimation](https://mlsanity.com/paper/2402.04878/) (2024)
- [A Coding-Theoretic Analysis of Hyperspherical Prototypical Learning Geometry](https://mlsanity.com/paper/2407.07664/) (2024)
- [Prior Availability in Industrial Visual Sim-to-Real: A Review of CAD-Guided and CAD-Unavailable Regimes](https://mlsanity.com/paper/2605.30581/) (2026)

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

- [CALIPER: Metric-Grounded Model-Free Recognition of Visually Similar Industrial Parts](https://mlsanity.com/paper/2609.17820/) (2026)

## Abstract (arXiv)

A scan supplies labeled images and a geometric reference. We separate their contributions in a recognizer whose scan-derived prototype matrix acts as a supervised head's fixed output layer. On T-LESS, HOPE and 18 self-collected industrial parts, we test real, random and exactly permuted prototypes, matched geometry-free classifiers, stronger appearance rules and paired background protocols. Across DINOv2-giant and MetaCLIP-H with real-background queries, the largest fused-accuracy advantage of the real prototypes over either control is one percentage point; larger differences favor controls, by up to 2.8 points in arm means. On HOPE with DINOv2-giant the head alone is 2.8 points above exact permutations (95% interval: 0.8-4.7); this advantage does not reach fusion and is not observed on MetaCLIP-H. On DINOv2-giant, matched logistic regression comes within 0.5 points of fusion on T-LESS and exceeds it on HOPE and the self-collected parts. Against white cutouts, real HOPE query backgrounds lower image-prototype accuracy by 43 points on DINOv2-giant and 13 on MetaCLIP-H. The audit separates prototype content, label supervision and query protocol.