Review — FCOS: Fully Convolutional One-Stage Object Detection

FCOS: Training Without the Use of Anchor Boxes

FCOS works by predicting a 4D vector (l, t, r, b) encoding the location of a bounding box at each foreground pixel
FCOS Demo:


1. FCOS: Network Architecture

FCOS: Network Architecture

1.1. Notations, Inputs, Outputs

1.2. Loss Function

1.3. Inference

2. Multi-level Prediction with FPN for FCOS

2.1. Multi-level Prediction with FPN

2.2. Center-ness for FCOS

Center-ness is computed by Eq. (3) and decays from 1 to 0 as the location deviates from the center of the object.

3. Ablation Study

3.1. Multi-level Prediction with FPN

3.2. Center-ness

Ablation study for the proposed center-ness branch on minival split

3.3. Other Improvements

FCOS vs. RetinaNet on the minival split with ResNet-50-FPN as the backbone

4. SOTA Comparison

4.1. MS COCO Test-Dev Split

FCOS vs. other state-of-the-art two-stage or one-stage detectors (single-model and single-scale results)

4.2. Extensions on Region Proposal Networks

FCOS as Region Proposal Networks vs. RPNs with FPN

PhD, Researcher. I share what I've learnt and done. :) My LinkedIn:, My Paper Reading List: