# Deformable Part Models are Convolutional Neural Networks

arXiv: https://arxiv.org/abs/1409.5403 · HTML: https://mlsanity.com/paper/1409.5403/

## Key facts

- Cited by 35 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 18 Sep 2014.

## Built on by (most cited first, 30 of 35)

- [CoTracker: It is Better to Track Together](https://mlsanity.com/paper/2307.07635/) (2023)
- [Understanding Robustness of Transformers for Image Classification](https://mlsanity.com/paper/2103.14586/) (2021)
- [Fine-Grained Image Analysis with Deep Learning: A Survey](https://mlsanity.com/paper/2111.06119/) (2021)
- [Deformable ProtoPNet: An Interpretable Image Classifier Using Deformable Prototypes](https://mlsanity.com/paper/2111.15000/) (2021)
- [Sequential End-to-end Network for Efficient Person Search](https://mlsanity.com/paper/2103.10148/) (2021)
- [Improving Weakly-supervised Object Localization via Causal Intervention](https://mlsanity.com/paper/2104.10351/) (2021)
- [Surround-View Vision-based 3D Detection for Autonomous Driving: A Survey](https://mlsanity.com/paper/2302.06650/) (2023)
- [UniPhy: Learning a Unified Constitutive Model for Inverse Physics Simulation](https://mlsanity.com/paper/2505.16971/) (2025)
- [Part-Based Models Improve Adversarial Robustness](https://mlsanity.com/paper/2209.09117/) (2022)
- [Vision transformers in domain adaptation and domain generalization: a study of robustness](https://mlsanity.com/paper/2404.04452/) (2024)
- [Domain Adaptive Person Search](https://mlsanity.com/paper/2207.11898/) (2022)
- [PartImageNet++ Dataset: Scaling up Part-based Models for Robust Recognition](https://mlsanity.com/paper/2407.10918/) (2024)
- [Distance-Aware Occlusion Detection with Focused Attention](https://mlsanity.com/paper/2208.11122/) (2022)
- [Spatiotemporal Deformable Scene Graphs for Complex Activity Detection](https://mlsanity.com/paper/2104.08194/) (2021)
- [Med-Query: Steerable Parsing of 9-DoF Medical Anatomies with Query Embedding](https://mlsanity.com/paper/2212.02014/) (2022)
- [Mamba Capsule Routing Towards Part-Whole Relational Camouflaged Object Detection](https://mlsanity.com/paper/2410.03987/) (2024)
- [AI perspectives in Smart Cities and Communities to enable road vehicle automation and smart traffic control](https://mlsanity.com/paper/2104.03150/) (2021)
- [Resource-Efficient Invariant Networks: Exponential Gains by Unrolled Optimization](https://mlsanity.com/paper/2203.05006/) (2022)
- [Sampling-based Uncertainty Estimation for an Instance Segmentation Network](https://mlsanity.com/paper/2305.14977/) (2023)
- [DP-Net: Learning Discriminative Parts for image recognition](https://mlsanity.com/paper/2404.15037/) (2024)
- [Learning Compositional Representations for Effective Low-Shot Generalization](https://mlsanity.com/paper/2204.08090/) (2022)
- [PS-ARM: An End-to-End Attention-aware Relation Mixer Network for Person Search](https://mlsanity.com/paper/2210.03433/) (2022)
- [DDAM-PS: Diligent Domain Adaptive Mixer for Person Search](https://mlsanity.com/paper/2310.20706/) (2023)
- [PARMESAN: Parameter-Free Memory Search and Transduction for Dense Prediction Tasks](https://mlsanity.com/paper/2403.11743/) (2024)
- [Fast One-Stage Unsupervised Domain Adaptive Person Search](https://mlsanity.com/paper/2405.02832/) (2024)
- [Sparse-shot Learning with Exclusive Cross-Entropy for Extremely Many Localisations](https://mlsanity.com/paper/2104.10425/) (2021)
- [Fine-grained Few-shot Recognition by Deep Object Parsing](https://mlsanity.com/paper/2207.07110/) (2022)
- [Preventing Errors in Person Detection: A Part-Based Self-Monitoring Framework](https://mlsanity.com/paper/2307.04533/) (2023)
- [Occlusion Aware Student Emotion Recognition based on Facial Action Unit Detection](https://mlsanity.com/paper/2307.09465/) (2023)
- [Domain Adaptive Person Search via GAN-based Scene Synthesis for Cross-scene Videos](https://mlsanity.com/paper/2308.04322/) (2023)

## Abstract (arXiv)

Deformable part models (DPMs) and convolutional neural networks (CNNs) are two widely used tools for visual recognition. They are typically viewed as distinct approaches: DPMs are graphical models (Markov random fields), while CNNs are "black-box" non-linear classifiers. In this paper, we show that a DPM can be formulated as a CNN, thus providing a novel synthesis of the two ideas. Our construction involves unrolling the DPM inference algorithm and mapping each step to an equivalent (and at times novel) CNN layer. From this perspective, it becomes natural to replace the standard image features used in DPM with a learned feature extractor. We call the resulting model DeepPyramid DPM and experimentally validate it on PASCAL VOC. DeepPyramid DPM significantly outperforms DPMs based on histograms of oriented gradients features (HOG) and slightly outperforms a comparable version of the recently introduced R-CNN detection system, while running an order of magnitude faster.