Explicit Shape Encoding for Real-Time Instance Segmentation
arXiv:1908.04067
The paper introduces ESE‑Seg, a top‑down instance segmentation framework that encodes object shapes explicitly using an inner‑center radius signature and Chebyshev polynomial fitting, enabling segmentation speeds comparable to object detection.
Abstract
In this paper, we propose a novel top-down instance segmentation framework based on explicit shape encoding, named \textbf{ESE-Seg}. It largely reduces the computational consumption of the instance segmentation by explicitly decoding the multiple object shapes with tensor operations, thus performs the instance segmentation at almost the same speed as the object detection. ESE-Seg is based on a novel shape signature Inner-center Radius (IR), Chebyshev polynomial fitting and the strong modern object detectors. ESE-Seg with YOLOv3 outperforms the Mask R-CNN on Pascal VOC 2012 at mAP$^r$@0.5 while 7 times faster.
to appear in ICCV2019