PSRR-MaxpoolNMS: Pyramid Shifted MaxpoolNMS with Relationship Recovery

TL;DR

This paper introduces PSRR-MaxpoolNMS, a parallelizable method that replaces GreedyNMS in object detection, achieving faster speeds and comparable accuracy through a novel relationship recovery and pyramid shifted approach.

Contribution

The paper presents PSRR-MaxpoolNMS, a new fully parallelizable NMS method that improves accuracy and speed over existing MaxpoolNMS and can replace GreedyNMS at all detection stages.

Findings

Outperforms MaxpoolNMS significantly in accuracy.

Faster than GreedyNMS with similar accuracy.

Enables hardware acceleration for NMS processes.

Abstract

Non-maximum Suppression (NMS) is an essential postprocessing step in modern convolutional neural networks for object detection. Unlike convolutions which are inherently parallel, the de-facto standard for NMS, namely GreedyNMS, cannot be easily parallelized and thus could be the performance bottleneck in convolutional object detection pipelines. MaxpoolNMS is introduced as a parallelizable alternative to GreedyNMS, which in turn enables faster speed than GreedyNMS at comparable accuracy. However, MaxpoolNMS is only capable of replacing the GreedyNMS at the first stage of two-stage detectors like Faster-RCNN. There is a significant drop in accuracy when applying MaxpoolNMS at the final detection stage, due to the fact that MaxpoolNMS fails to approximate GreedyNMS precisely in terms of bounding box selection. In this paper, we propose a general, parallelizable and configurable approach PSRR-MaxpoolNMS, to completely replace GreedyNMS at all stages in all detectors. By introducing a simple Relationship Recovery module and a Pyramid Shifted MaxpoolNMS module, our PSRR-MaxpoolNMS is able to approximate GreedyNMS more precisely than MaxpoolNMS. Comprehensive experiments show that our approach outperforms MaxpoolNMS by a large margin, and it is proven faster than GreedyNMS with comparable accuracy. For the first time, PSRR-MaxpoolNMS provides a fully parallelizable solution for customized hardware design, which can be reused for accelerating NMS everywhere.