Computation Reallocation for Object Detection

TL;DR

This paper introduces CR-NAS, a neural architecture search method that optimizes computation resource allocation in object detection backbones, leading to improved accuracy without extra computational cost.

Contribution

The paper proposes a novel hierarchical search method for dynamic computation reallocation across feature resolutions and spatial positions in object detection models.

Findings

CR-NAS improves COCO AP by 1.9% with CR-ResNet50.

CR-NAS enhances CR-MobileNetV2 by 1.7% COCO AP.

The discovered models are transferable to other datasets and tasks.

Abstract

The allocation of computation resources in the backbone is a crucial issue in object detection. However, classification allocation pattern is usually adopted directly to object detector, which is proved to be sub-optimal. In order to reallocate the engaged computation resources in a more efficient way, we present CR-NAS (Computation Reallocation Neural Architecture Search) that can learn computation reallocation strategies across different feature resolution and spatial position diectly on the target detection dataset. A two-level reallocation space is proposed for both stage and spatial reallocation. A novel hierarchical search procedure is adopted to cope with the complex search space. We apply CR-NAS to multiple backbones and achieve consistent improvements. Our CR-ResNet50 and CR-MobileNetV2 outperforms the baseline by 1.9% and 1.7% COCO AP respectively without any additional computation budget. The models discovered by CR-NAS can be equiped to other powerful detection neck/head and be easily transferred to other dataset, e.g. PASCAL VOC, and other vision tasks, e.g. instance segmentation. Our CR-NAS can be used as a plugin to improve the performance of various networks, which is demanding.