320 likes | 537 Views
Salient Object Detection for Searched Web Images via Global Saliency. Peng Wang 1 Jingdong Wang 2 Gang Zeng 1 Jie Feng 1 Hongbin Zha 1 Shipeng Li 2 1 Key Laboratory on Machine Perception, Peking University 2 Microsoft Research Asia. Outline. Introduction Web Image Database
E N D
Salient Object Detection for Searched Web Images via Global Saliency Peng Wang1Jingdong Wang2 Gang Zeng1Jie Feng1 Hongbin Zha1Shipeng Li2 1Key Laboratory on Machine Perception, Peking University 2Microsoft Research Asia
Outline • Introduction • Web Image Database • Proposed Algorithm • Experiments • Conclusion and Discussion
Introduction • Want to detect the existence and the location of salient objects for thumbnail images. • Use a learning approach, random forest for solution • Sliding window-based method • Segmentation-based method
Web Image Database • Searched 1100 queries and downloaded 400 thumbnail images for each query. • Each image I(x) is assigned a corresponding label vector y = (o, t, l, b, r). • Bounding box distribution
Proposed Algorithm • Input & output space • training features • output labels • mapping function
Proposed Algorithm • Creating features
Proposed Algorithm • Creating features • Multi-scale Contract (MC) • Center-Surrounding Histogram (CSH) • Region-based Contrast (RC) • Color Spatial Distribution (CSD) With a number of saliency maps Sk(x) in which k = 1,…,K.
Proposed Algorithm • Creating features • Combining strategies • Stack Partition each Sk(x) into N = p × p blocks in a grid layout, then extract the average value in each block.
Proposed Algorithm • Creating features • Combining strategies • SumUp Apply a non-linear combination of multiple saliency values as: λk : the weight of the kth saliency map
Proposed Algorithm • Detecting object existence via classification • Feed the features into the random forest classifier to learn the mapping gc. • Down-sampling the majority class.
Proposed Algorithm • Translation and scale invariance feature • Perform a rectification on the saliency map Sall(x). • Fit the two dimensional Gaussian function • Find A, µ,Σ by minimizing the objective • Translate the image center to the position µ = (µx, µy)T • the range of coordinate x: [µx − λσx, µx + λσx]
Proposed Algorithm • Localizing object via regression • Learning the posterior distribution p(w|f ) through regression. • Construct the multiple partition {Pz}Zz=1 • Combined through averaging • estimate the position
Experiments • Databases • MSRA image set B with images resized into 130 × 130 • Web image database
Experiments • Classification evaluation
Experiments • Regression evaluation • Training parameters • Number of training image • Minimum size of each node
Experiments • Regression evaluation • Effectiveness of features
Experiments • Comparison with other approaches • use two measurements • region-based measurement Overlap score
Experiments • Comparison with other approaches • use two measurements • edge-based measurement Bounding box Boundary Displacement Error (BDE)
Experiments • Comparison with other approaches
Experiments • Comparison with other approaches
Experiments • Comparison with other approaches
Conclusion and Discussion • Presented a large labeled web image database. • A supervised scheme that judges the existence of and predicts the location of the salient object in thumbnail images. • Exploits random forest and global saliency with features created from the saliency maps combining information of multiple channels.