Weakly Supervised Localization and Learning with Generic Knowledge

By Thomas Deselaers, Bogdan Alexe, Vittorio Ferrari. 2012

306 citations recorded by the source corpus.

Subfields: Artificial Intelligence, 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 Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Webly Supervised Learning of Convolutional Networks (2015, methodology): “The outline of our approach is shown in Figure 2.”
Learning Open-World Object Proposals without Learning to Classify (2021, background): “They have become the integral pre-processing steps for many computer vision systems, including object detection [33,41,42,45,59,74], segmentation [3,8,13], object discovery [12, 16, 47], weakly supervised object detection [5, 21, 53], visual tracking [35, 63], content-aware retargeting [51], etc.”
Extreme Clicking for Efficient Object Annotation (2017, background): “This setting is very cheap in terms of annotation time, but it produces lower quality object detectors, typically performing at only about half the level of accuracy achieved by training from bounding boxes [5, 11, 14, 55, 67].”
Object-Proposal Evaluation Protocol is ‘Gameable’ (2015, background): “Notice that the former definition has an emphasis on object discovery [27,28,30], while the latter definition emphasises on the ultimate performance of a detection pipeline.”
Self-taught object localization with deep networks (2014, methodology): “or methods: detectors trained on our automatically-generated bounding boxes perform nearly on par with detectors learned from ground-truth annotations. Most weakly-supervised object detection methods [9, 8, 24, 27, 3] aim at jointly learning and inferring both the class and the position of the objec”
Efficient Image and Video Co-localization with Frank-Wolfe Algorithm (2014, methodology): “We also show results compared to state-of-the-art co-localization methods in Table 2 for the PASCAL07-all dataset [14], which does not consider the “bird”, “car”, “cat”, “cow”, “dog”, and “sheep” classes.”
W2F: A Weakly-Supervised to Fully-Supervised Framework for Object Detection (2018, methodology): “Also, the correct location (CorLoc) [7] metric is used to evaluate the localization accuracy of our model on the training set.”
C-MIL: Continuation Multiple Instance Learning for Weakly Supervised Object Detection (2019, methodology): “The used Correct Localization (CorLoc) metric [36] is the percentage of images for which the region of highest object score has at least 0.”
Discover and Learn New Objects from Documentaries (2017, methodology): “The timeconsuming annotation process in object detection can be sidestepped by weakly supervised learning [2, 3, 7, 9, 13, 30, 33, 36, 40, 41, 42, 43, 44].”
Bayesian Joint Modelling for Object Localisation in Weakly Labelled Images (2015, methodology): “The existing (discriminative) line of work focusing on WSOL [11], [13], [42], [43] has not generally exploited unlabelled data, and cannot”
Co-localization in Real-World Images (2014, methodology): “Co-localization shares the same type of output as weakly supervised localization [9, 21, 22, 27], where we draw bounding boxes around objects without any strong supervision.”
C-MIDN: Coupled Multiple Instance Detection Network With Segmentation Guidance for Weakly Supervised Object Detection (2019, methodology): “For evaluation, we use two kinds of measurements: 1) Average Precision (AP) and the mean of AP (mAP) on the test set, following the standard PAS-CAL VOC protocol; 2) CorLoc [8] on the trainval set to evaluate the localization accuracy.”
Detector discovery in the wild: Joint multiple instance and representation learning (2014, methodology): “[10] uses auxiliary data to learn generic objectness information just as an initial step, but doesn’t optimize jointly for weakly labeled data.”
Advanced Deep-Learning Techniques for Salient and Category-Specific Object Detection: A Survey (2018, methodology): “To alleviate this problem, weakly supervised object detection approaches [3], [13], [76], [99] have received wide interest in recent years.”
Learning to Separate Object Sounds by Watching Unlabeled Video (2018, background): “, with a CRF [15] or multiple instance learning (MIL) [1,9,13,71].”
Harnessing Lab Knowledge for Real-World Action Recognition (2014, background): “In this way, it is more effective in learning a new object class from cluttered training images (Deselaers et al. 2012).”