Skip to main navigation Skip to search Skip to main content

Guided refine-head for object detection

  • Lingyun Zeng
  • , You Song*
  • , Wenhai Wang
  • *Corresponding author for this work
  • Beihang University
  • Nanjing University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In recent years, multi-stage detectors improve the accuracy of object detection to a new level. However, due to multiple stages, these methods typically fall short in the inference speed. To alleviate this problem, we propose a novel object detector—Guided Refine-Head, which is made up of a newly proposed detection network called Refine-Head and a knowledge-distillation-like loss function. Refine-Head is a two-stage detector, and thus Refine-Head has faster inference speed than multi-stage detectors. Nonetheless, Refine-Head is able to predict bounding boxes for incremental IoU thresholds like a multi-stage detector. In addition, we use knowledge-distillation-like loss function to guide the training process of Refine-Head. Therefore, besides fast inference speed, the proposed Guided Refine-Head also has competitive accuracy. Abundant ablation studies and comparative experiments on MS-COCO 2017 validate the superiority of the proposed Guided Refine-Head. It is worth noting that Guided Refine-Head achieves the AP of 38.0% at 10.4 FPS, surpassing Faster R-CNN by 1.8% at the similar speed.

Original languageEnglish
Title of host publicationMultiMedia Modeling - 26th International Conference, MMM 2020, Proceedings
EditorsWen-Huang Cheng, Junmo Kim, Jung-Woo Choi, Wei-Ta Chu, Peng Cui, Min-Chun Hu, Wesley De Neve
PublisherSpringer
Pages203-214
Number of pages12
ISBN (Print)9783030377304
DOIs
StatePublished - 2020
Event26th International Conference on MultiMedia Modeling, MMM 2020 - Daejeon, Korea, Republic of
Duration: 5 Jan 20208 Jan 2020

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11961 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference26th International Conference on MultiMedia Modeling, MMM 2020
Country/TerritoryKorea, Republic of
CityDaejeon
Period5/01/208/01/20

Keywords

  • Knowledge distillation
  • Object detection
  • Two-stage detector

Fingerprint

Dive into the research topics of 'Guided refine-head for object detection'. Together they form a unique fingerprint.

Cite this