# Zoom Out-and-In Network with Map Attention Decision for Region Proposal and Object Detection

arXiv: https://arxiv.org/abs/1709.04347 · HTML: https://mlsanity.com/paper/1709.04347/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 13 Sep 2017.

## Built on by (most cited first, 13 of 13)

- [Learning Open-World Object Proposals without Learning to Classify](https://mlsanity.com/paper/2108.06753/) (2021)
- [CAT: LoCalization and IdentificAtion Cascade Detection Transformer for Open-World Object Detection](https://mlsanity.com/paper/2301.01970/) (2023)
- [Regional Attention with Architecture-Rebuilt 3D Network for RGB-D Gesture Recognition](https://mlsanity.com/paper/2102.05348/) (2021)
- [Towards Balanced Learning for Instance Recognition](https://mlsanity.com/paper/2108.10175/) (2021)
- [CE-FPN: Enhancing Channel Information for Object Detection](https://mlsanity.com/paper/2103.10643/) (2021)
- [Indoor Obstacle Discovery on Reflective Ground via Monocular Camera](https://mlsanity.com/paper/2401.01445/) (2024)
- [Survey: Exploiting Data Redundancy for Optimization of Deep Learning](https://mlsanity.com/paper/2208.13363/) (2022)
- [An Attention Module for Convolutional Neural Networks](https://mlsanity.com/paper/2108.08205/) (2021)
- [A survey on bias in machine learning research](https://mlsanity.com/paper/2308.11254/) (2023)
- [Targeted Data Augmentation for bias mitigation](https://mlsanity.com/paper/2308.11386/) (2023)
- [Improving Open-World Object Localization by Discovering Background](https://mlsanity.com/paper/2504.17626/) (2025)
- [Augmenting Anchors by the Detector Itself](https://mlsanity.com/paper/2105.14086/) (2021)
- [Small Object Detection Model with Spatial Laplacian Pyramid Attention and Multi-Scale Features Enhancement in Aerial Images](https://mlsanity.com/paper/2602.23031/) (2026)

## Abstract (arXiv)

In this paper, we propose a zoom-out-and-in network for generating object proposals. A key observation is that it is difficult to classify anchors of different sizes with the same set of features. Anchors of different sizes should be placed accordingly based on different depth within a network: smaller boxes on high-resolution layers with a smaller stride while larger boxes on low-resolution counterparts with a larger stride. Inspired by the conv/deconv structure, we fully leverage the low-level local details and high-level regional semantics from two feature map streams, which are complimentary to each other, to identify the objectness in an image. A map attention decision (MAD) unit is further proposed to aggressively search for neuron activations among two streams and attend the most contributive ones on the feature learning of the final loss. The unit serves as a decisionmaker to adaptively activate maps along certain channels with the solely purpose of optimizing the overall training loss. One advantage of MAD is that the learned weights enforced on each feature channel is predicted on-the-fly based on the input context, which is more suitable than the fixed enforcement of a convolutional kernel. Experimental results on three datasets, including PASCAL VOC 2007, ImageNet DET, MS COCO, demonstrate the effectiveness of our proposed algorithm over other state-of-the-arts, in terms of average recall (AR) for region proposal and average precision (AP) for object detection.