# AutoFocus: Efficient Multi-Scale Inference

## TL;DR

AutoFocus introduces an efficient multi-scale inference method for object detection that selectively processes regions likely to contain small objects, significantly reducing computation while maintaining high accuracy.

## Contribution

It proposes a novel coarse-to-fine inference approach using FocusPixels and FocusChips, enabling faster detection with minimal accuracy loss compared to traditional multi-scale methods.

## Key findings

- Achieves 47.9% mAP on COCO test-dev at 6.4 images/sec.
- Reduces pixels processed by 5X with only 1% mAP drop.
- Outperforms RetinaNet in mAP while maintaining similar speed.

## Abstract

This paper describes AutoFocus, an efficient multi-scale inference algorithm for deep-learning based object detectors. Instead of processing an entire image pyramid, AutoFocus adopts a coarse to fine approach and only processes regions which are likely to contain small objects at finer scales. This is achieved by predicting category agnostic segmentation maps for small objects at coarser scales, called FocusPixels. FocusPixels can be predicted with high recall, and in many cases, they only cover a small fraction of the entire image. To make efficient use of FocusPixels, an algorithm is proposed which generates compact rectangular FocusChips which enclose FocusPixels. The detector is only applied inside FocusChips, which reduces computation while processing finer scales. Different types of error can arise when detections from FocusChips of multiple scales are combined, hence techniques to correct them are proposed. AutoFocus obtains an mAP of 47.9% (68.3% at 50% overlap) on the COCO test-dev set while processing 6.4 images per second on a Titan X (Pascal) GPU. This is 2.5X faster than our multi-scale baseline detector and matches its mAP. The number of pixels processed in the pyramid can be reduced by 5X with a 1% drop in mAP. AutoFocus obtains more than 10% mAP gain compared to RetinaNet but runs at the same speed with the same ResNet-101 backbone.