# Find Beauty in the Rare: Contrastive Composition Feature Clustering for Nontrivial Cropping Box Regression

arXiv: https://arxiv.org/abs/2302.08662 · HTML: https://mlsanity.com/paper/2302.08662/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 49 references that are in the map.
- First posted on arXiv 17 Feb 2023.

## Builds on (most cited first, 22 of 22)

- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [MobileNetV2: Inverted Residuals and Linear Bottlenecks](https://mlsanity.com/paper/1801.04381/) (2018)
- [Unsupervised Feature Learning via Non-Parametric Instance-level Discrimination](https://mlsanity.com/paper/1805.01978/) (2018)
- [Understanding Contrastive Representation Learning through Alignment and Uniformity on the Hypersphere](https://mlsanity.com/paper/2005.10242/) (2020)
- [Decoupling Representation and Classifier for Long-Tailed Recognition](https://mlsanity.com/paper/1910.09217/) (2019)
- [Invariant Information Clustering for Unsupervised Image Classification and Segmentation](https://mlsanity.com/paper/1807.06653/) (2018)
- [Discriminative Unsupervised Feature Learning with Exemplar Convolutional Neural Networks](https://mlsanity.com/paper/1406.6909/) (2014)
- [Unsupervised Embedding Learning via Invariant and Spreading Instance Feature](https://mlsanity.com/paper/1904.03436/) (2019)
- [Local Aggregation for Unsupervised Learning of Visual Embeddings](https://mlsanity.com/paper/1903.12355/) (2019)
- [Multi-task Self-Supervised Visual Learning](https://mlsanity.com/paper/1708.07860/) (2017)
- [Delving into Deep Imbalanced Regression](https://mlsanity.com/paper/2102.09554/) (2021)
- [A Comprehensive Survey on Deep Clustering: Taxonomy, Challenges, and Future Directions](https://mlsanity.com/paper/2206.07579/) (2022)
- [Side-Aware Boundary Localization for More Precise Object Detection](https://mlsanity.com/paper/1912.04260/) (2019)
- [3D Instances as 1D Kernels](https://mlsanity.com/paper/2207.07372/) (2022)
- [Reliable and Efficient Image Cropping: A Grid Anchor based Approach](https://mlsanity.com/paper/1904.04441/) (2019)
- [Quantitative Analysis of Automatic Image Cropping Algorithms: A Dataset and Comparative Study](https://mlsanity.com/paper/1701.01480/) (2017)
- [Grid Anchor based Image Cropping: A New Benchmark and An Efficient Model](https://mlsanity.com/paper/1909.08989/) (2019)
- [Image Cropping with Composition and Saliency Aware Aesthetic Score Map](https://mlsanity.com/paper/1911.10492/) (2019)
- [Strongly Augmented Contrastive Clustering](https://mlsanity.com/paper/2206.00380/) (2022)
- [Design What You Desire: Icon Generation from Orthogonal Application and Theme Labels](https://mlsanity.com/paper/2208.00439/) (2022)
- [Vision Transformer for Contrastive Clustering](https://mlsanity.com/paper/2206.12925/) (2022)

## Built on by (most cited first, 7 of 7)

- [Exposure Completing for Temporally Consistent Neural High Dynamic Range Video Rendering](https://mlsanity.com/paper/2407.13309/) (2024)
- [Instance Consistency Regularization for Semi-Supervised 3D Instance Segmentation](https://mlsanity.com/paper/2406.16776/) (2024)
- [AesCrop: Aesthetic-driven Cropping Guided by Composition](https://mlsanity.com/paper/2510.22528/) (2025)
- [CROP: Expert-Aligned Image Cropping via Compositional Reasoning and Optimizing Preference](https://mlsanity.com/paper/2605.12545/) (2026)
- [R4DSG: Relative 4D Scene Graph Memory for Object-Centric Question Answering in Long Egocentric Video](https://mlsanity.com/paper/2608.11017/) (2026)
- [Pseudo-Labeling by Multi-Policy Viewfinder Network for Image Cropping](https://mlsanity.com/paper/2407.01971/) (2024)
- [ShotCrop3: Cropping Human-Centric Images into Cinematic Triple-Shot Compositions](https://mlsanity.com/paper/2606.05635/) (2026)

## Abstract (arXiv)

Automatic image cropping algorithms aim to recompose images like human-being photographers by generating the cropping boxes with improved composition quality. Cropping box regression approaches learn the beauty of composition from annotated cropping boxes. However, the bias of annotations leads to quasi-trivial recomposing results, which has an obvious tendency to the average location of training samples. The crux of this predicament is that the task is naively treated as a box regression problem, where rare samples might be dominated by normal samples, and the composition patterns of rare samples are not well exploited. Observing that similar composition patterns tend to be shared by the cropping boundaries annotated nearly, we argue to find the beauty of composition from the rare samples by clustering the samples with similar cropping boundary annotations, ie, similar composition patterns. We propose a novel Contrastive Composition Clustering (C2C) to regularize the composition features by contrasting dynamically established similar and dissimilar pairs. In this way, common composition patterns of multiple images can be better summarized, which especially benefits the rare samples and endows our model with better generalizability to render nontrivial results. Extensive experimental results show the superiority of our model compared with prior arts. We also illustrate the philosophy of our design with an interesting analytical visualization.