# Prototypical Priors: From Improving Classification to Zero-Shot Learning

arXiv: https://arxiv.org/abs/1512.01192 · HTML: https://mlsanity.com/paper/1512.01192/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 Dec 2015.

## Built on by (most cited first, 14 of 14)

- [Rethinking Semantic Segmentation: A Prototype View](https://mlsanity.com/paper/2203.15102/) (2022)
- [Not All Features Matter: Enhancing Few-shot CLIP with Adaptive Prior Refinement](https://mlsanity.com/paper/2304.01195/) (2023)
- [Visual Recognition with Deep Nearest Centroids](https://mlsanity.com/paper/2209.07383/) (2022)
- [Taming Cross-Domain Representation Variance in Federated Prototype Learning with Heterogeneous Data Domains](https://mlsanity.com/paper/2403.09048/) (2024)
- [COARSE3D: Class-Prototypes for Contrastive Learning in Weakly-Supervised 3D Point Cloud Segmentation](https://mlsanity.com/paper/2210.01784/) (2022)
- [Prototype-based Dataset Comparison](https://mlsanity.com/paper/2309.02401/) (2023)
- [ProMotion: Prototypes As Motion Learners](https://mlsanity.com/paper/2406.04999/) (2024)
- [Prototypical Transformer as Unified Motion Learners](https://mlsanity.com/paper/2406.01559/) (2024)
- [Multi-Scale Grouped Prototypes for Interpretable Semantic Segmentation](https://mlsanity.com/paper/2409.09497/) (2024)
- [Hyperspherical Classification with Dynamic Label-to-Prototype Assignment](https://mlsanity.com/paper/2403.16937/) (2024)
- [Proto-FG3D: Prototype-based Interpretable Fine-Grained 3D Shape Classification](https://mlsanity.com/paper/2505.17666/) (2025)
- [Introducing the structural bases of typicality effects in deep learning](https://mlsanity.com/paper/2107.03279/) (2021)
- [An Enhanced Federated Prototype Learning Method under Domain Shift](https://mlsanity.com/paper/2409.18578/) (2024)
- [DPL: Decoupled Prototype Learning for Enhancing Robustness of Vision-Language Transformers to Missing Modalities](https://mlsanity.com/paper/2505.08283/) (2025)

## Abstract (arXiv)

Recent works on zero-shot learning make use of side information such as visual attributes or natural language semantics to define the relations between output visual classes and then use these relationships to draw inference on new unseen classes at test time. In a novel extension to this idea, we propose the use of visual prototypical concepts as side information. For most real-world visual object categories, it may be difficult to establish a unique prototype. However, in cases such as traffic signs, brand logos, flags, and even natural language characters, these prototypical templates are available and can be leveraged for an improved recognition performance. The present work proposes a way to incorporate this prototypical information in a deep learning framework. Using prototypes as prior information, the deepnet pipeline learns the input image projections into the prototypical embedding space subject to minimization of the final classification loss. Based on our experiments with two different datasets of traffic signs and brand logos, prototypical embeddings incorporated in a conventional convolutional neural network improve the recognition performance. Recognition accuracy on the Belga logo dataset is especially noteworthy and establishes a new state-of-the-art. In zero-shot learning scenarios, the same system can be directly deployed to draw inference on unseen classes by simply adding the prototypical information for these new classes at test time. Thus, unlike earlier approaches, testing on seen and unseen classes is handled using the same pipeline, and the system can be tuned for a trade-off of seen and unseen class performance as per task requirement. Comparison with one of the latest works in the zero-shot learning domain yields top results on the two datasets mentioned above.