Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

MobileSal: Extremely Efficient RGB-D Salient Object Detection

About

The high computational cost of neural networks has prevented recent successes in RGB-D salient object detection (SOD) from benefiting real-world applications. Hence, this paper introduces a novel network, MobileSal, which focuses on efficient RGB-D SOD using mobile networks for deep feature extraction. However, mobile networks are less powerful in feature representation than cumbersome networks. To this end, we observe that the depth information of color images can strengthen the feature representation related to SOD if leveraged properly. Therefore, we propose an implicit depth restoration (IDR) technique to strengthen the mobile networks' feature representation capability for RGB-D SOD. IDR is only adopted in the training phase and is omitted during testing, so it is computationally free. Besides, we propose compact pyramid refinement (CPR) for efficient multi-level feature aggregation to derive salient objects with clear boundaries. With IDR and CPR incorporated, MobileSal performs favorably against state-of-the-art methods on six challenging RGB-D SOD datasets with much faster speed (450fps for the input size of 320 $\times$ 320) and fewer parameters (6.5M). The code is released at https://mmcheng.net/mobilesal.

Yu-Huan Wu, Yun Liu, Jun Xu, Jia-Wang Bian, Yu-Chao Gu, Ming-Ming Cheng• 2020

Related benchmarks

TaskDatasetResultRank
Salient Object DetectionVT5000
S-Measure0.754
50
Salient Object DetectionVT821
S-Measure0.654
36
Salient Object DetectionVT1000
Fm (F-measure)0.784
19
Salient Object DetectionUVT 2000
Fm57.1
18
Salient Object Detectionun-VT5000
Fm70.5
18
Salient Object Detectionun-VT1000
Fm79.6
18
Salient Object Detectionun-VT821
Fm63.6
18
Salient Object DetectionUVT20K
F-measure (Fm)0.744
17
Showing 8 of 8 rows

Other info

Follow for update