CVOct 6, 2020

Joint COCO and Mapillary Workshop at ICCV 2019: COCO Instance Segmentation Challenge Track

arXiv:2010.02475v15 citations
Originality Synthesis-oriented
AI Analysis

This work addresses the problem of improving instance segmentation accuracy for computer vision applications, representing an incremental advancement in performance on a specific benchmark.

The paper tackles object detection and instance segmentation by presenting MegDetV2, a two-pass system that first detects instances and then obtains segmentation, achieving 61.0 mAP for detection and 53.1 mAP for instance segmentation on the COCO-2019 test-dev dataset, surpassing previous winning results by 5.0 and 4.2 respectively.

In this report, we present our object detection/instance segmentation system, MegDetV2, which works in a two-pass fashion, first to detect instances then to obtain segmentation. Our baseline detector is mainly built on a new designed RPN, called RPN++. On the COCO-2019 detection/instance-segmentation test-dev dataset, our system achieves 61.0/53.1 mAP, which surpassed our 2018 winning results by 5.0/4.2 respectively. We achieve the best results in COCO Challenge 2019 and 2020.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes