Published on Tue Nov 21 2017

S4Net: Single Stage Salient-Instance Segmentation

Ruochen Fan, Ming-Ming Cheng, Qibin Hou, Tai-Jiang Mu, Jingdong Wang, Shi-Min Hu

We consider an interesting problem-salient instance segmentation in this paper. Other than producing bounding boxes, our network also outputs high-quality instance-level segments. We evaluate our approach on a publicly available benchmark and show that it outperforms other solutions.

0
0
0
Abstract

We consider an interesting problem-salient instance segmentation in this paper. Other than producing bounding boxes, our network also outputs high-quality instance-level segments. Taking into account the category-independent property of each target, we design a single stage salient instance segmentation framework, with a novel segmentation branch. Our new branch regards not only local context inside each detection window but also its surrounding context, enabling us to distinguish the instances in the same scope even with obstruction. Our network is end-to-end trainable and runs at a fast speed (40 fps when processing an image with resolution 320x320). We evaluate our approach on a publicly available benchmark and show that it outperforms other alternative solutions. We also provide a thorough analysis of the design choices to help readers better understand the functions of each part of our network. The source code can be found at \url{https://github.com/RuochenFan/S4Net}.

Wed Aug 19 2020
Machine Learning
Salient Instance Segmentation with Region and Box-level Annotations
Salient instance segmentation is a new challenging task that received widespread attention in the saliency detection area. The proposed method achieves a mask AP of 58.3% that outperforms the mainstream state-of-the-art methods.
0
0
0
Wed Apr 12 2017
Computer Vision
Instance-Level Salient Object Segmentation
Image saliency detection has recently witnessed rapid progress due to deep convolutional neural networks. None of the existing methods is able to identify object instances in the detected salient regions. We propose amultiscale saliency refinement network, which generates high-quality salient region masks and salient object
0
0
0
Fri Aug 28 2020
Machine Learning
Regularized Densely-connected Pyramid Network for Salient Instance Segmentation
The code is available at https://github.com/yuhuan-wu/RDPNet. Extensive experiments on popular benchmarks demonstrate that our design significantly outperforms existing competitors by 6.3%.
0
0
0
Sun Sep 29 2019
Computer Vision
Salient Instance Segmentation via Subitizing and Clustering
The goal of salient region detection is to identify the regions of an image that attract the most attention. Many methods have achieved state-of-the-art performance levels on this task. In this paper, we propose a multitask densely connected neural network (MDNN) to segment salient instances in an image.
0
0
0
Sun Aug 15 2021
Computer Vision
SOTR: Segmenting Objects with Transformers
Most recent transformer-based models show impressive performance on vision tasks. Segmenting Objects with TRansformers (SOTR) simplifies the segmentation pipeline. SOTR can effectively extract lower-level feature representations and capture long-range context.
5
21
129
Wed Oct 28 2020
Machine Learning
Class-Agnostic Segmentation Loss and Its Application to Salient Object Detection and Segmentation
With CAS loss the class descriptors are learned during training of the network. CAS loss clusters regions with similar appearance together together in a weakly-supervised manner. For low-fidelity training data (incorrect class label) class-agnostic segmentation loss outperforms the state-of-the-
0
0
0