CPM R-CNN: Calibrating Point-guided Misalignment in Object Detection

03/07/2020
by   Bin Zhu, et al.
0

In object detection, offset-guided and point-guided regression dominate anchor-based and anchor-free method separately. Recently, point-guided approach is introduced to anchor-based method. However, we observe points predicted by this way are misaligned with matched region of proposals and score of localization, causing a notable gap in performance. In this paper, we propose CPM R-CNN which contains three efficient modules to optimize anchor-based point-guided method. According to sufficient evaluations on the COCO dataset, CPM R-CNN is demonstrated efficient to improve the localization accuracy by calibrating mentioned misalignment. Compared with Faster R-CNN and Grid R-CNN based on ResNet-101 with FPN, our approach can substantially improve detection mAP by 3.3 best model achieves improvement by a large margin to 49.9 Code and models will be publicly available.

READ FULL TEXT

page 6

page 12

research
12/04/2019

Multiple Anchor Learning for Visual Object Detection

Classification and localization are two pillars of visual object detecto...
research
05/11/2020

Scope Head for Accurate Localization in Object Detection

Existing anchor-based and anchor-free object detectors in multi-stage or...
research
11/29/2018

Grid R-CNN

This paper proposes a novel object detection framework named Grid R-CNN,...
research
04/24/2018

An Anchor-Free Region Proposal Network for Faster R-CNN based Text Detection Approaches

The anchor mechanism of Faster R-CNN and SSD framework is considered not...
research
01/10/2019

Region Proposal by Guided Anchoring

Region anchors are the cornerstone of modern object detection techniques...
research
06/13/2019

Grid R-CNN Plus: Faster and Better

Grid R-CNN is a well-performed objection detection framework. It transfo...
research
12/24/2015

Adaptive Object Detection Using Adjacency and Zoom Prediction

State-of-the-art object detection systems rely on an accurate set of reg...

Please sign up or login with your details

Forgot password? Click here to reset