Measuring the Objectness of Image Windows

By Bogdan Alexe, Thomas Deselaers, Vittorio Ferrari. 2011

1,286 citations recorded by the source corpus.

We present a generic objectness measure, quantifying how likely it is for an image window to contain an object of any class. We explicitly train it to distinguish objects with a well-defined boundary in space, such as cows and telephones, from amorphous background elements, such as grass and road. The measure combines in a Bayesian framework several image cues measuring characteristics of objects, such as appearing different from their surroundings and having a closed boundary. These include an innovative cue to measure the closed boundary characteristic. In experiments on the challenging PASCAL VOC 07 dataset, we show this new cue to outperform a state-of-the-art saliency measure, and the combined objectness measure to perform better than any cue alone. We also compare to interest point operators, a HOG detector, and three recent works aiming at automatic object segmentation. Finally, we present two applications of objectness. In the first, we sample a small numberof windows according to their objectness probability and give an algorithm to employ them as location priors for modern class-specific object detectors. As we show experimentally, this greatly reduces the number of windows evaluated by the expensive class-specific model. In the second application, we use objectness as a complementary score in addition to the class-specific model, which leads to fewer false positives. As shown in several recent papers, objectness can act as a valuable focus of attention mechanism in many other applications operating on image windows, including weakly supervised learning of object categories, unsupervised pixelwise segmentation, and object tracking in video. Computing objectness is very efficient and takes only about 4 sec. per image.

Subfields: Computer Vision and Pattern Recognition

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

Citations in the corpus

Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks (2015, neutral): “, objectness in windows [24], EdgeBoxes [6]).”
ImageNet Large Scale Visual Recognition Challenge (2014, methodology): “To quantify clutter we employ the objectness measure of (Alexe et al., 2012), which is a class-generic object detector evaluating how likely a window in the image contains a coherent object (of any class) as opposed to background (sky, water, grass).”
Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation (2013, methodology): “We followed the same protocol of submitting test results to the ILSVRC2013 evaluation server only twice, once with and once without bounding-box regression.”
Training Region-Based Object Detectors with Online Hard Example Mining (2016, methodology): “R-CNN, in contrast, uses region proposals [1, 2, 3, 4, 6, 10, 18, 32, 34], a method that was made popular by the selective search algorithm [32].”
Learning to Refine Object Segments (2016, methodology): “We apply our approach in the context of object proposal generation [32–38].”
Learning to Segment Object Candidates (2015, background): “These approaches usually fall into three categories: (1) objectness scoring [1, 34], in which proposals are extracted by measuring the objectness score of bounding boxes, (2) seed segmentation [14, 16, 17], where models start with multiple seed regions and generate separate foreground-background seg”
What Makes for Effective Detection Proposals? (2015, methodology): “Likewise, object proposals considerably reduce computation compared to the dense (sliding window) detection framework by generating candidate proposals that may contain objects.”
COCO-Stuff: Thing and Stuff Classes in Context (2016, background): “rform signiﬁcantly above chance at 59.1%, 60.2% and 67.8%, respectively. The relative image position cue performs suprisingly well at 87.7%, conﬁrming the positional bias observed in many prior works [32,52,61,28,4,9]. The more complex objectness measure performs best at 90.6%. Using all cues we ach”
Object-Proposal Evaluation Protocol is ‘Gameable’ (2015, background): “In early works [2, 4, 6], the emphasis was on category independent object proposals, where the goal is to identify instances of all objects in the image irrespective of their category.”
A MultiPath Network for Object Detection (2016, background): “Object Proposals: When originally introduced, object proposals were based on lowlevel grouping cues, edges, and superpixels [1, 2, 16, 31, 33].”
Learning visual groups from co-occurrences in space and time (2015, background): “…showing up in the tasks of edge/contour detection Canny (1986); Arbelaez et al. (2011); Isola et al. (2014) , (semantic) segmentation Shi & Malik (2000); Malisiewicz & Efros (2007), and object proposals Alexe et al. (2012); Zitnick & Doll ´ ar (2014); Krahnenbuhl & Koltun (2015), among others.”
HCP: A Flexible CNN Framework for Multi-Label Image Classification (2014, background): “methods [1], [2], [4], [9], [38], [44] have been proposed to generate a set of hypotheses to cover all independent objects in a given image.”
What's the Point: Semantic Segmentation with Point Supervision (2015, methodology): “These scores are obtained by using the pretrained objectness model from the released code of [2].”
Attend Refine Repeat: Active Box Proposal Generation via In-Out Localization (2016, background): “In the objectness scoring paradigm, a large set of image boxes is ranked according to how likely it is for each image box to tightly enclose an object — regardless of its category — and then this set is post-processed with a non-maximum-suppression step and truncated to yield the ﬁnal set of object…”