FCOS: Fully Convolutional One-Stage Object Detection

2 Apr 2019Zhi TianChunhua ShenHao ChenTong He

We propose a fully convolutional one-stage object detector (FCOS) to solve object detection in a per-pixel prediction fashion, analogue to semantic segmentation. Almost all state-of-the-art object detectors such as RetinaNet, SSD, YOLOv3, and Faster R-CNN rely on pre-defined anchor boxes... (read more)

PDF Abstract

Evaluation results from the paper

Task Dataset Model Metric name Metric value Global rank Compare
Object Detection COCO FCOS_ResNeXt-101-32x8d-FPN Bounding Box AP 42.1 # 16
Object Detection COCO FCOS_ResNet-101-FPN Bounding Box AP 41.0 # 19