A Simple Pooling-Based Design for Real-Time Salient Object Detection

Abstract

We solve the problem of salient object detection by investigating how to expand the role of pooling in convolutional neural networks. Based on the U-shape architecture, we first build a global guidance module (GGM) upon the bottom-up pathway, aiming at providing layers at different feature levels the location information of potential salient objects. We further design a feature aggregation module (FAM) to make the coarse-level semantic information well fused with the fine-level features from the top-down pathway. By adding FAMs after the fusion operations in the topdown pathway, coarse-level features from the GGM can be seamlessly merged with features at various scales. These two pooling-based modules allow the high-level semantic features to be progressively refined, yielding detail enriched saliency maps. Experiment results show that our proposed approach can more accurately locate the salient objects with sharpened details and hence substantially improve the performance compared to the previous state-of-the-arts. Our approach is fast as well and can run at a speed of more than 30 FPS when processing a 300 × 400 image. Code can be found at http://mmcheng.net/poolnet/.

Res2Net: A New Multi-Scale Backbone Architecture

“…Multi-scale feature representations of CNNs are of great importance to a number of vision tasks including object detection [43], face analysis [4], [41], edge detection [37], semantic segmentation [6], salient object detection [34], [65], and skeleton detection [67], boosting the model performance of those fields.…”

RGB-T Salient Object Detection via Fusing Multi-Level CNN Features

“…To validate the proposed RGB-T salient detection model, we compare our model with 10 SOTA methods, which are further divided into three types, i.e., (1)RGB salient object detection methods: PoolNet [39], R3Net [40], and CPDNet [41]; (2) RGB-D salient object detection methods: AFNet [45], TSAA [46], PDNet [47], and SSRC [48]; and (3) RGB-T salient object detection methods: MFSR [28], GCL [49], and MRCM [27]. For fair comparisons, we modify these RGB and RGB-D salient object detection methods for RGB-T saliency detection.…”

“…Recently, to extract more sophisticated features, tremendous deep learning based saliency detectors have been proposed [19]- [25], [39]- [41], and achieved substantially better performance than those previous methods. For example, Lee et al [23] proposed to first encode low-level distance map and high-level sematic features of deep CNNs to form a new feature vector, and then evaluate saliency by a multi-level fully connected neural network classifier.…”

“…Zhang et al [25] introduced a generic aggregating multi-level convolutional feature framework for salient object detection, which first integrated multi-level feature maps into multiple resolutions and then adaptively learned to combine these feature maps at each resolution to predict the saliency maps. In [39], two pooling based modules, i.e., a global guidance module (GGM) and a feature aggregation module (FAM), aided to improve the performance for salient object detection. A novel recurrent residual refinement network (R3Net) equipped with residual refinement blocks (RRBs) was presented in [40] to detect salient regions from an input image more accurately.…”

Deeper and Mixed Supervision for Salient Object Detection in Automated Surface Inspection

“…We compare the effectiveness of the DMS model with 16 saliency methods, including 12 traditional saliency algorithms (ITTI [40], LC [41], SR [42], AC [43], FT [44], MSS [45], PHOT [46], HC [47], RC [47], SF [48], BMS [49], and MBP [50]), and 8 deep learning methods (U-Net [51], FCN [19], R3Net [24], DSS [33], PiCANet [52], BASNet [29], PoolNet [53], and EGNet [26]). We implement the traditional saliency algorithm through the toolbox provided in [16].…”