---
title: Centroid-Level Representations
url: https://www.emergentmind.com/topics/centroid-level-representations
type: topic
---

# Centroid-Level Representations

Centroid-level representations refer to the use of explicit cluster centroids—mean, prototype, or aggregate vectors—to summarize structural, semantic, or statistical properties of groups of data points at various abstraction layers. These centroids can represent classes, clusters, objects, concepts, or even geometric or probabilistic aggregations across a variety of data modalities (images, text, 3D shapes, uncertain data), supporting tasks such as classification, clustering, retrieval, instance association, and efficient neural inference. The methodologies for computing, updating, and employing centroids differ by application domain, learning paradigm, and theoretical underpinnings, including vector space, non-Euclidean manifolds, or probabilistic models.

## 1. Mathematical Formalisms and Centroid Construction

The mathematical definition of centroids varies across contexts:

- **Vector Spaces:** In standard settings, a centroid $c_k$ for class $k$ is the arithmetic mean over all sample embeddings,
  $$
  c_k = \frac{1}{N_k} \sum_{i=1}^{N_k} f(x_i),
  $$
  where $f(x_i)$ is the feature embedding and $N_k$ the cardinality [2104.13643, 2102.08606].
  
- **Probabilistic/Uncertain Data:** The U-centroid is a random variable induced by the distributions of all objects in the cluster. For cluster $C=\{o_1,...,o_{|C|}\}$, with each $o_i$ a random variable, the U-centroid is $X_c = \frac{1}{|C|} \sum_i Y_i$, with $Y_i \sim f_i$ independently [1203.6401].

- **Non-Euclidean Manifolds:** For hyperbolic embeddings, centroids are not vector means but points constructed via iterated Möbius addition and scaling or constructive algorithms based on geodesic midpoints and recursion (e.g., LFC, LBC, LAC, BTC) [2211.04462].

- **Object-centric Representations:** In vision, object-centric centroids $c_n$ are 2D positions calculated as a weighted pixel average within an object mask, i.e.,
  $$
  c_n = \frac{1}{\sum_p M_n(p)}\sum_p p\, M_n(p),
  $$
  for hard mask $M_n$ [2511.14072].

- **Structural Shape Averaging:** The centroid in elastic mean contour frameworks involves a closed-form for optimal translation, coupling parameterization and centroid determination via double energy minimization [1901.02823].

These formalisms underlie practical algorithms for efficient learning and inference across domains.

## 2. Centroid-Level Representations in Supervised and Unsupervised Learning

- **Class Prototypes:** In supervised settings, centroids serve as class prototypes, supporting non-parametric classification, contrastive learning, and incremental learning by providing class reference points or adaptation anchors [2412.18219, 2104.13643, 2110.10048].

- **Clustering:** Unsupervised algorithms such as Agg-Var clustering in concept learning [1911.00155], partitional clustering under uncertainty [1203.6401], and centroid attention layers in transformers [2102.08606] all depend on iterative or online updates of centroids defined by either proximity constraints, distributional aggregation, or differentiable soft assignments.

- **Contrastive Learning:** EMA-updated class centroids are crucial in long-tailed recognition, where interpolated embeddings are explicitly trained to retrieve multiple centroids, regularizing both head and tail representations [2110.10048].

- **One-Class and Anomaly Detection:** Centroid-level representations are employed to define "bona fide" clusters—using an adaptive centroid updated via a running weighted average—which a one-class loss then compacts while forcing outliers away [2406.16716].

## 3. Object-Centric and Spatial Centroid Representations

In structured perception, centroids are tightly linked to semantic or physical object entities:

- **Token Compression and Object-centric LVLMs:** CORE forms object tokens by pooling representations within segmentation masks to their centroids, yielding compact yet semantically aligned visual tokens. Centroid-guided sorting preserves spatial order for efficient LLM fusion [2511.14072].

- **Human Pose and Instance Segmentation:** VISUALCENT uses keypoint-centric and mask-centric offsets to dynamically estimate per-person and per-instance centroids, enabling robust joint segmentation in crowded or occluded scenes [2504.19032].

- **3D Shape Descriptors:** The SGC descriptor encodes local geometry by computing centroids and point densities within voxelized grids referenced by local object-centric frames, yielding robust partial-to-partial shape fingerprints [1612.08408].

## 4. Computational Procedures and Algorithmic Variations

Centroid computation is adapted to suit application constraints:

| Method/class                  | Update Rule or Computation                       | Key Features                    |
|-------------------------------|--------------------------------------------------|----------------------------------|
| Mean/EMA (Euclidean)          | Simple mean or $m c^k + (1-m)z_i$               | O(1) per class, for large $K$   |
| Agg-Var clustering            | Online thresholded update, create-or-merge       | Variable cluster sizes          |
| Hyperbolic centroid schemes   | Recursive Möbius addition, tree-based aggregation| Non-Euclidean structure         |
| Centroid attention (transformer) | Single-step gradient on clustering loss           | O(NM) vs O(N²), abstraction     |
| Mask-based centroiding        | Weighted pixel average in object mask            | Semantically grounded           |
| U-centroid for uncertainty    | Induced random variable, closed-form for $J(C)$  | Captures mean & variance        |

Each procedure can be made efficiently scalable (e.g., online centroid tracking [1911.00155], KNN-masked attention [2102.08606], adaptive merging in incremental learning [2412.18219], descriptor-graphs for fast partial matching [1612.08408]).

## 5. Theoretical Properties, Advantages, and Limitations

- **Interpretablility:** Centroid-level representations offer explicitly interpretable geometric or semantic prototypes, enabling direct model inspection, cluster merging, or semantic editing [1911.00155, 2511.14072].

- **Scalability and Efficiency:** By reducing the number of representative vectors, centroids support efficient storage, retrieval, and inference, critical for large-scale or resource-constrained deployments [2104.13643, 2511.14072].

- **Robustness:** Aggregation mitigates outlier influence, label noise, and internal variance, supporting improved accuracy, especially in challenging scenarios such as low-sample tail classes, major modality shift, or partial data [2110.10048, 2406.16716, 1612.08408].

- **Limitations:** Single centroids can obscure intra-class multimodality; centroid-based metrics in non-Euclidean or highly non-linear manifolds may misalign with actual task-relevant similarity [1911.00155, 2211.04462]. Threshold and parameter selection often remains empirical, and heuristic assignment rules (e.g., voting, reciprocal distance) are not learned end-to-end.

## 6. Applications and Empirical Impact

Centroid-level representations have demonstrated state-of-the-art (SOTA) or substantially improved empirical results in:

- **Image retrieval and re-identification:** Reducing search space and boosting mAP and top-1 accuracy over instance-based approaches [2104.13643].
- **Long-tailed and low-shot recognition:** Dramatic tail-class accuracy improvements due to centroid-driven contrastive training [2110.10048].
- **Class-incremental learning (CIL):** Preserving prototype alignment via affine centroid mapping yields both superior accuracy and $O(1)$ inference time, exceeding SOTA methods under scalability constraints [2412.18219].
- **Efficient LVLM inference:** CORE, under object-centric token merging with centroid-based sorting, achieves SOTA compression with minimal drop in task performance, and maintains spatial semantics at token-level [2511.14072].
- **Human analysis, 3D vision, and geometric computing:** VISUALCENT and SGC produce robust, efficiently computed centroid-level signatures enabling real-time segmentation and partial shape matching, respectively [2504.19032, 1612.08408].
- **Non-Euclidean text classification:** Hyperbolic centroid schemes, especially LBC and LAC, can match or exceed Euclidean baselines in hierarchical datasets, under appropriately constructed embeddings [2211.04462].

## 7. Extensions, Open Questions, and Future Perspectives

There is continued development in:

- **Centroid schemes for non-Euclidean/complex manifolds:** Advancing Riemannian mean computation (Fréchet/Karcher means), optimal transport barycenters, and centroids for graphs or structured objects [2211.04462].
- **Multimodality and hierarchical structures:** Multi-centroid or sub-class prototyping for classes with rich internal structure, and centroids for hierarchical aggregation [2104.13643, 1911.00155].
- **Adaptive centroid construction:** Online, distributed, and adaptive centroid updates for continual and federated learning scenarios [2412.18219].
- **Integrating centroids with deep architecture compression or token pruning:** Using centroid-guided abstraction within transformers and LVLMs to simultaneously optimize efficiency and semantic retention [2511.14072, 2102.08606].
- **Task-specific centroid measures:** Aligning centroid distances with non-linear task constraints (e.g., complex geometry, uncertainty) and exploring theory for centroid-induced isoperimetric, generalization, and robustness properties [1902.10614, 1203.6401].

The explicit use of centroid-level representations—spanning clustering, prototypical learning, transformer bottlenecking, and geometric aggregation—constitutes a unifying and versatile paradigm with demonstrated impact across contemporary AI research domains.

Source: https://www.emergentmind.com/topics/centroid-level-representations