Explicit Shape Encoding for Real-Time Instance Segmentation

By Wenqiang Xu, Haiyang Wang, Fubo Qi, Cewu Lu. 2019

113 citations recorded by the source corpus.

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.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

K-Net: Towards Unified Image Segmentation (2021, background): “Some studies explore polar [58] representation, contour [45], and explicit shape representation [60] of instance masks.”
Forest R-CNN: Large-Vocabulary Long-Tailed Object Detection and Instance Segmentation (2020, methodology): “or Faster R-CNN [32]. On the basis of the two-stage fashion, [20, 21, 25] further propose several framework modifications so as to yield better performance. On the other hand, single-stage approaches [1, 3, 10, 43], aiming at a faster inference speed, adopt a more straightforward way to directly gen”
CNN-GCN Aggregation Enabled Boundary Regression for Biomedical Image Segmentation (2020, methodology): “Instead of dense pixels classiﬁcation, recent works [24,3,25] exploited object contours for eﬃcient segmentation and achieved comparable performance with pixel-based segmentation methods.”
Interactive Two-Stream Decoder for Accurate and Fast Saliency Detection (2020, methodology): “In the U-shape network, features from the encoder are skip-connected to the features with the same spatial sizes in the decoder, where this simple connection has been demonstrated its power in various pixel-wise predictions [14, 39, 43].”
PolarMask: Single Shot Instance Segmentation With Polar Representation (2019, methodology): “Concurrent to our work is the work of ESESeg [30], which also employs the polar coordinate to model instances.”
MapTR: Structured Modeling and Learning for Online Vectorized HD Map Construction (2022, background): “Another line of work related to MapTR is contour-based 2D instance segmentation (Zhu et al., 2022; Xie et al., 2020; Xu et al., 2019; Liu et al., 2021c).”
SipMask: Spatial Information Preservation for Fast Image and Video Instance Segmentation (2020, methodology): “ving spatial information in bounding-box. The spatial coecients split mask prediction into multiple sub-mask predictions, leading to improved mask quality. two-stage [21,31,8,23,17] and single-stage [37,13,47,2,42,36] methods, based on the underlying detection framework. Two-stage methods typically ”