Fuzzy-NMS: Improving 3D Object Detection with Fuzzy Classification in NMS

TL;DR

Fuzzy-NMS introduces a fuzzy classification-based filtering method into NMS to enhance 3D object detection accuracy, especially for small objects, without retraining or significant inference time increase.

Contribution

The paper proposes a novel Fuzzy-NMS module that refines candidate bounding boxes using fuzzy classification, improving detection accuracy in 3D object detection frameworks.

Findings

Significant accuracy improvements on KITTI and Waymo benchmarks.

Enhanced detection of small objects like pedestrians and bicycles.

Fuzzy-NMS is a plug-and-play module with no retraining required.

Abstract

Non-maximum suppression (NMS) is an essential post-processing module used in many 3D object detection frameworks to remove overlapping candidate bounding boxes. However, an overreliance on classification scores and difficulties in determining appropriate thresholds can affect the resulting accuracy directly. To address these issues, we introduce fuzzy learning into NMS and propose a novel generalized Fuzzy-NMS module to achieve finer candidate bounding box filtering. The proposed Fuzzy-NMS module combines the volume and clustering density of candidate bounding boxes, refining them with a fuzzy classification method and optimizing the appropriate suppression thresholds to reduce uncertainty in the NMS process. Adequate validation experiments are conducted using the mainstream KITTI and large-scale Waymo 3D object detection benchmarks. The results of these tests demonstrate the proposed Fuzzy-NMS module can improve the accuracy of numerous recently NMS-based detectors significantly, including PointPillars, PV-RCNN, and IA-SSD, etc. This effect is particularly evident for small objects such as pedestrians and bicycles. As a plug-and-play module, Fuzzy-NMS does not need to be retrained and produces no obvious increases in inference time.