Saliency filters: Contrast based filtering for salient region detection

By Federico Perazzi, Philipp Krahenbuhl, Yael Pritch, Alexander Sorkine-Hornung. 2012 IEEE Conference on Computer Vision and Pattern Recognition, 2012

2,181 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/7abac1f397a76bc06d7508ab4108462786395c0f

How researchers see this paper

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

In Computer Vision and Pattern Recognition

1. Ming-Hsuan Yang
2. Luc Van Gool
3. Ming-Ming Cheng
4. Ling Shao
5. Dacheng Tao
6. Huchuan Lu
7. Xiaogang Wang
8. Yi Yang

Citations in the corpus

• A Stagewise Refinement Model for Detecting Salient Objects in Images (2017, methodology): “Complementary to the PR curve, we also report the Mean Absolute Error (MAE) which is calculated as the average pixelwise absolute difference between the binary groundtruth G and the saliency map S adopted by [37],”
• Salient Object Detection via Structured Matrix Decomposition (2017, methodology): “Given a continuous saliency map S and the binary ground truth G, MAE is defined as the mean absolute difference between S andG : MAE = mean(|S−G|) [50].”
• Surface defect saliency of magnetic tile (2018, methodology): “The 14models in the toolbox are: ITTI [30] and BMS [14] based on visual-attention mechanism; FT [28], LC [46] and HC [47] based on global color rarity; AC [34] and MSS [29] based on local color rarity; SR [48], Rudinac [49] and PHOT [36] based on frequency domain; RC [47], SF [50] and GMR [26] based”
• RGBD Salient Object Detection: A Benchmark and Algorithms (2014, methodology): “We select 12 state-of-the-art RGB saliency detection approaches, including the top 4 models ranked in the survey[6]: SVO[9], CA[14], CB[20], and RC[10], and 8 recently developed prominent methods: SF[34], LR[40], HS[44], MR[45], PCA[32], TD[37], GC[11], and DSR[28].”
• Saliency detection by multi-context deep learning (2015, methodology): “A large number of approaches [63, 52, 40, 39, 32, 35, 60, 57, 56, 47, 41, 31, 27, 25, 24, 23, 11, 44, 17, 8, 13, 1, 21] are proposed to capture different saliency cues.”
• Learning to Look Around: Intelligently Exploring Unseen Environments for Unknown Tasks (2017, background): “Saliency and attention: Previous work studies the question of “where to look” to prioritize portions of already captured image/video data, so as to reserve computation for the most salient regions or block out distractors [1, 5, 8, 23, 41, 47, 52, 62, 71], or to predict the gaze or preference of a h”
• New Contour Cue-Based Hybrid Sparse Learning for Salient Object Detection (2019, methodology): “…we denote as BMS [10], CA [52], CB [35], CGVS [53], CRF [15], DSR [20], FES [54], FT [16], GMR [30], GR [36], GS [28], HS [55], LC [56], LEGS [42], LMLC [41], LPS [57], MBS (MBS+) [45], MC [29], MNP [58], PSO [59], RC [17], SEG [60], SeR [61], SF [49], ST [62], and SVO [34], respectively.”
• UIU-Net: U-Net in U-Net for Infrared Small Object Detection (2022, background): “Currently, there are some other works in the field of saliency object detection, e.g., Contrast based filtering [47], Context-aware saliency [48], spectral residual method [49], etc., which can also achieve the Currently, detecting unknown scenes or objects is still a challenging and noteworthy…”
• DHSNet: Deep Hierarchical Saliency Network for Salient Object Detection (2016, background): “Compactness prior [12] advocates that salient object regions are compact and perceptually homogeneous elements.”
• Deeply Supervised Salient Object Detection with Short Connections (2016, methodology): “We use three universally-agreed, standard metrics (see also [23], [23], [41], [58]) to evaluate our model including precision-recall curves, F-measure, and the mean absolute error (MAE).”
• Salient object detection via bootstrap learning (2015, methodology): “Left to right: (a) input (b) GS SP [34] (c) wCO [43] (d) LRMR [30] (e) GMR [38] (f) DSR [23] (g) XL13 [36] (h) HS [37] (i) RC-J [9] (j) GC [10] (k) SF [28] (l) Ours (m) wCO-bootstrapped (n) ground truth.”
• Towards Explanation of DNN-based Prediction with Guided Feature Inversion (2018, methodology): “Our method is compared with six state-of-the-art salient object detection methods, including Signature saliency (SS) [18], Simulation (SIM) [29], Region-based contrast (RC) [3], Frequency-tuned (FT) [1], Saliency filters (SF) [30], Bootstrap learning (BL) [41].”
• Non-local Deep Features for Salient Object Detection (2017, methodology): “MAE [37] is computed as the average pixel-wise absolute difference between the estimated saliency map S and its corresponding ground truth L , where W and H is the width and height of a given image.”
• UberNet: Training a Universal Convolutional Neural Network for Low-, Mid-, and High-Level Vision Using Diverse Datasets and Limited Memory (2016, result): “We compare to some classic methods [18, 45, 80] as well as more recent ones [58, 83, 102, 112] that typically rely on deep learning.”
• SwinNet: Swin Transformer Drives Edge-Aware RGB-D and RGB-T Salient Object Detection (2022, background): “They are the precision-recall (PR) curve [83], S-measure [84], F-measure [85], E-measure [86] and mean absolute error (MAE) [87].”
• Hierarchical Saliency Detection (2013, methodology): “[23] measured the variance of spatial distribution for each color.”
• Co-localization in Real-World Images (2014, methodology): “We compute an off-the-shelf saliency map for each image [6, 23], and for each box we compute the average saliency within the box, weighted by the size of the box, and stack these values into the nb dimensional vector mbox to obtain a linear term that penalizes less salient boxes:”
• The Secrets of Salient Object Segmentation (2014, methodology): “Previous literature of salient object segmentation [1, 9, 21] often claim a big margin of F-measures over all fixation algorithms.”
• Progressive Attention Guided Recurrent Network for Salient Object Detection (2018, methodology): “Salient Object Detection methods can be categorized as conventional low-level hand-crafted features based [20, 4, 33, 15, 9, 38, 23, 21] and Convolutional Neural Networks driven [25, 13, 37, 14, 12, 16, 27, 18, 7, 35, 36, 28] approaches.”