Local Background Enclosure for RGB-D Salient Object Detection - Supplementary Results
نویسندگان
چکیده
The purpose of this supplementary material is to examine in detail the contributions of our proposed Local Background Enclosure (LBE) feature. A comparison of LBE with the contrast based depth features used in state-of-the-art salient object detection systems is presented. The LBE feature is compared with the raw depth features ACSD [1], DC [3] and a signed version of DC denoted SDC on the RGBD1000 [2] and NJUDS2000 [1] datasets. We then qualitatively assess the use of the prior application and Grabcut refinement stages of our saliency system to enhance the LBE saliency map. Please find attached the outputs of the tested salient object detection systems on each dataset. The saliency maps for ACSD [1], LMH [2], and GP [3] were obtained by running code from the author’s websites. Note that the saliency maps have been resized to conserve space.
منابع مشابه
Salient Object Detection via Bootstrap Learning Supplementary Materials
In Figure 2(a), we provide the Precision and Recall (P-R) curves of the proposed bootstrap learning algorithm when each of the three features (RGB, CIELab and LBP) is removed, respectively. These results show that each feature contributes to detect salient objects. In addition, we show the P-R curve of the proposed method in comparison with those of single-scale methods in Figure 2(b), which de...
متن کاملPDNet: Prior-model Guided Depth-enhanced Network for Salient Object Detection
Fully convolutional neural networks (FCNs) have shown outstanding performance in many computer vision tasks including salient object detection. However, there still remains two issues needed to be addressed in deep learning based saliency detection. One is the lack of tremendous amount of annotated data to train a network. The other is the lack of robustness for extracting salient objects in im...
متن کاملObject Recognition based on Local Steering Kernel and SVM
The proposed method is to recognize objects based on application of Local Steering Kernels (LSK) as Descriptors to the image patches. In order to represent the local properties of the images, patch is to be extracted where the variations occur in an image. To find the interest point, Wavelet based Salient Point detector is used. Local Steering Kernel is then applied to the resultant pixels, in ...
متن کاملAutomatic Salient Object Detection for Panoramic Images Using Region Growing and Fixation Prediction Model
Saliency detection aims to detect the most attractive objects in images, which has been widely used as a foundation for various multimedia applications. In this paper, we propose a novel salient object detection algorithm for RGB-D images using center-dark channel prior. First, we generate an initial saliency map based on a color saliency map and a depth saliency map of a given RGB-D image. The...
متن کاملSaliency-aware Real-time Volumetric Fusion for Object Reconstruction
We present a real-time approach for acquiring 3D objects with high fidelity using hand-held consumer-level RGB-D scanning devices. Existing real-time reconstruction methods typically do not take the point of interest into account, and thus might fail to produce clean reconstruction results of desired objects due to distracting objects or backgrounds. In addition, any changes in background durin...
متن کامل