# ViCrop-Det: Spatial Attention Entropy Guided Cropping for Training-Free Small-Object Detection

arXiv: https://arxiv.org/abs/2604.26806 · HTML: https://mlsanity.com/paper/2604.26806/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 26 of its 32 references that are in the map.
- First posted on arXiv 29 Apr 2026.

## Builds on (most cited first, 26 of 26)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [End-to-End Object Detection with Transformers](https://mlsanity.com/paper/2005.12872/) (2020)
- [Deformable DETR: Deformable Transformers for End-to-End Object Detection](https://mlsanity.com/paper/2010.04159/) (2020)
- [Cascade R-CNN: Delving into High Quality Object Detection](https://mlsanity.com/paper/1712.00726/) (2017)
- [DynamicViT: Efficient Vision Transformers with Dynamic Token Sparsification](https://mlsanity.com/paper/2106.02034/) (2021)
- [DOTA: A Large-scale Dataset for Object Detection in Aerial Images](https://mlsanity.com/paper/1711.10398/) (2017)
- [Learning Continuous Image Representation with Local Implicit Image Function](https://mlsanity.com/paper/2012.09161/) (2020)
- [DETRs Beat YOLOs on Real-time Object Detection](https://mlsanity.com/paper/2304.08069/) (2023)
- [Localizing Objects with Self-Supervised Transformers and no Labels](https://mlsanity.com/paper/2109.14279/) (2021)
- [AdaViT: Adaptive Vision Transformers for Efficient Image Recognition](https://mlsanity.com/paper/2111.15668/) (2021)
- [Dynamic Head: Unifying Object Detection Heads with Attentions](https://mlsanity.com/paper/2106.08322/) (2021)
- [Instances as Queries](https://mlsanity.com/paper/2105.01928/) (2021)
- [Not All Images are Worth 16x16 Words: Dynamic Transformers for Efficient Image Recognition](https://mlsanity.com/paper/2105.15075/) (2021)
- [An Analysis of Scale Invariance in Object Detection - SNIP](https://mlsanity.com/paper/1711.08189/) (2017)
- [FeatUp: A Model-Agnostic Framework for Features at Any Resolution](https://mlsanity.com/paper/2403.10516/) (2024)
- [TokenCut: Segmenting Objects in Images and Videos with Self-supervised Transformer and Normalized Cut](https://mlsanity.com/paper/2209.00383/) (2022)
- [Slicing Aided Hyper Inference and Fine-tuning for Small Object Detection](https://mlsanity.com/paper/2202.06934/) (2022)
- [Interpreting CNNs via Decision Trees](https://mlsanity.com/paper/1802.00121/) (2018)
- [SNIPER: Efficient Multi-Scale Training](https://mlsanity.com/paper/1805.09300/) (2018)
- [Clustered Object Detection in Aerial Images](https://mlsanity.com/paper/1904.08008/) (2019)
- [LSKNet: A Foundation Lightweight Backbone for Remote Sensing](https://mlsanity.com/paper/2403.11735/) (2024)
- [Less is More: Focus Attention for Efficient DETR](https://mlsanity.com/paper/2307.12612/) (2023)
- [DQ-DETR: DETR with Dynamic Query for Tiny Object Detection](https://mlsanity.com/paper/2404.03507/) (2024)
- [Deep Interpretable Classification and Weakly-Supervised Segmentation of Histology Images via Max-Min Uncertainty](https://mlsanity.com/paper/2011.07221/) (2020)
- [AttEntropy: On the Generalization Ability of Supervised Semantic Segmentation Transformers to New Objects in New Domains](https://mlsanity.com/paper/2212.14397/) (2022)

## Abstract (arXiv)

Transformer-based architectures have established a dominant paradigm in global semantic perception; however, they remain fundamentally constrained by the profound spatial heterogeneity inherent in natural images. Specifically, the imposition of a uniform global receptive field across regions of varying information density inevitably leads to local feature degradation, particularly in dense conflict zones populated by microscopic targets. To address this mechanistic limitation, we propose ViCrop-Det, a training-free inference framework that introduces adaptive spatial trust region shrinkage. Inspired by the use of attention entropy in anomaly segmentation, ViCrop-Det leverages the detection decoder's cross-attention distribution as an endogenous probe. By utilizing Spatial Attention Entropy (SAE) to heuristically evaluate local spatial ambiguity, the framework executes dynamic spatial routing, allocating a fixed computational budget exclusively to regions exhibiting both high target saliency and high cognitive uncertainty. By shrinking the spatial trust region and injecting high-frequency localized observations, ViCrop-Det actively resolves spatial ambiguity and recovers fine-grained features without requiring architectural modifications. Extensive evaluations on VisDrone and DOTA-v1.5 demonstrate that ViCrop-Det yields competitive performance enhancements, consistently adding +1-3 mAP@50 to RT-DETR-R50 and Deformable DETR with a marginal 20-23% latency overhead. On MS COCO, APS improves while APM/APL remains stable, indicating precise fine-scale refinement without compromising the global spatial prior. Under compute-matched settings, our adaptive routing strategy comprehensively surpasses uniform slicing baselines, achieving a highly optimized accuracy-speed trade-off.