POD: Practical Object Detection with Scale-Sensitive Network

TL;DR

This paper introduces a practical, scale-sensitive object detection network that explicitly models scale, improves accuracy on COCO, and is efficient for real-time applications with hardware acceleration support.

Contribution

The method predicts a global scale for each convolution filter, uses scale decomposition for fast deployment, and simplifies training without complex strategies.

Findings

Achieves 41.5 mAP on COCO test-dev with one-stage detector.

Outperforms baseline models by over 2 mAP points.

Supports hardware acceleration without extra operations.

Abstract

Scale-sensitive object detection remains a challenging task, where most of the existing methods could not learn it explicitly and are not robust to scale variance. In addition, the most existing methods are less efficient during training or slow during inference, which are not friendly to real-time applications. In this paper, we propose a practical object detection method with scale-sensitive network.Our method first predicts a global continuous scale ,which is shared by all position, for each convolution filter of each network stage. To effectively learn the scale, we average the spatial features and distill the scale from channels. For fast-deployment, we propose a scale decomposition method that transfers the robust fractional scale into combination of fixed integral scales for each convolution filter, which exploits the dilated convolution. We demonstrate it on one-stage and two-stage algorithms under different configurations. For practical applications, training of our method is of efficiency and simplicity which gets rid of complex data sampling or optimize strategy. During test-ing, the proposed method requires no extra operation and is very supportive of hardware acceleration like TensorRT and TVM. On the COCO test-dev, our model could achieve a 41.5 mAP on one-stage detector and 42.1 mAP on two-stage detectors based on ResNet-101, outperforming base-lines by 2.4 and 2.1 respectively without extra FLOPS.