# Structure-Regularized Attention for Deformable Object Representation

arXiv: https://arxiv.org/abs/2106.06672 · HTML: https://mlsanity.com/paper/2106.06672/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 43 references that are in the map.
- First posted on arXiv 12 Jun 2021.

## Builds on (most cited first, 23 of 23)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift](https://mlsanity.com/paper/1502.03167/) (2015)
- [MobileNetV2: Inverted Residuals and Linear Bottlenecks](https://mlsanity.com/paper/1801.04381/) (2018)
- [Squeeze-and-Excitation Networks](https://mlsanity.com/paper/1709.01507/) (2017)
- [Deep Learning Face Attributes in the Wild](https://mlsanity.com/paper/1411.7766/) (2014)
- [Non-local Neural Networks](https://mlsanity.com/paper/1711.07971/) (2017)
- [Self-Attention with Relative Position Representations](https://mlsanity.com/paper/1803.02155/) (2018)
- [Deep Image Prior](https://mlsanity.com/paper/1711.10925/) (2017)
- [CosFace: Large Margin Cosine Loss for Deep Face Recognition](https://mlsanity.com/paper/1801.09414/) (2018)
- [VGGFace2: A dataset for recognising faces across pose and age](https://mlsanity.com/paper/1710.08092/) (2017)
- [CCNet: Criss-Cross Attention for Semantic Segmentation](https://mlsanity.com/paper/1811.11721/) (2018)
- [Stand-Alone Self-Attention in Vision Models](https://mlsanity.com/paper/1906.05909/) (2019)
- [Beyond Part Models: Person Retrieval with Refined Part Pooling (and a Strong Convolutional Baseline)](https://mlsanity.com/paper/1711.09349/) (2017)
- [Learning Face Representation from Scratch](https://mlsanity.com/paper/1411.7923/) (2014)
- [Attention Augmented Convolutional Networks](https://mlsanity.com/paper/1904.09925/) (2019)
- [Cross-Age LFW: A Database for Studying Cross-Age Face Recognition in Unconstrained Environments](https://mlsanity.com/paper/1708.08197/) (2017)
- [Harmonious Attention Network for Person Re-Identification](https://mlsanity.com/paper/1802.08122/) (2018)
- [Expectation-Maximization Attention Networks for Semantic Segmentation](https://mlsanity.com/paper/1907.13426/) (2019)
- [Interaction-and-Aggregation Network for Person Re-identification](https://mlsanity.com/paper/1907.08435/) (2019)
- [Unsupervised learning of object landmarks by factorized spatial embeddings](https://mlsanity.com/paper/1705.02193/) (2017)
- [Visual Saliency Maps Can Apply to Facial Expression Recognition](https://mlsanity.com/paper/1811.04544/) (2018)

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

- [Learning Meta Representations for Agents in Multi-Agent Reinforcement Learning](https://mlsanity.com/paper/2108.12988/) (2021)

## Abstract (arXiv)

Capturing contextual dependencies has proven useful to improve the representational power of deep neural networks. Recent approaches that focus on modeling global context, such as self-attention and non-local operation, achieve this goal by enabling unconstrained pairwise interactions between elements. In this work, we consider learning representations for deformable objects which can benefit from context exploitation by modeling the structural dependencies that the data intrinsically possesses. To this end, we provide a novel structure-regularized attention mechanism, which formalizes feature interaction as structural factorization through the use of a pair of light-weight operations. The instantiated building blocks can be directly incorporated into modern convolutional neural networks, to boost the representational power in an efficient manner. Comprehensive studies on multiple tasks and empirical comparisons with modern attention mechanisms demonstrate the gains brought by our method in terms of both performance and model complexity. We further investigate its effect on feature representations, showing that our trained models can capture diversified representations characterizing object parts without resorting to extra supervision.