Multi-fold MIL Training for Weakly Supervised Object Localization

By R. G. Cinbis, Jakob Verbeek, Cordelia Schmid. 2014

241 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Evaluation & Benchmarks

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

In Evaluation & Benchmarks

Citations in the corpus

Fully Convolutional Multi-Class Multiple Instance Learning (2014, background): “MIL can reduce the need for bounding box annotations (Cinbis et al., 2014; Song et al., 2014), but it is rarely attempted for segmentation.”
Grounding of Textual Phrases in Images by Reconstruction (2015, background): “se are limited to small set of nouns. Object co-localization focuses on discovering and detecting an object in images or videos without any bounding box annotation, only from image/video level labels [3,6,24,31,39,40,50]. In [3] a structural SVM model is learned to detect objects from image-level su”
Self-taught object localization with deep networks (2014, methodology): “f images containing multiple objects or with context correlating with the object. Generalization to new datasets and novel classes. We show the ability of the proposed method to generalize to Classes [5] [31] STL CL aeroplane 56.6 80.1 79.0 bicycle 58.3 63.9 55.6 bird 28.4 51.5 48.8 boat 20.7 14.9 4”
Weakly Supervised Object Localization with Multi-Fold Multiple Instance Learning (2015, neutral): “Part of the material presented here appeared in [14].”
Unsupervised Object Discovery and Tracking in Video Collections (2015, background): “To reduce human effort and subjective biases in manual annotation, recent work has addressed the discovery and localization of objects from weakly-annotated or even unlabelled datasets [4, 5, 9, 30, 32].”
Y Qualification Main task Y Return N N Y User qualified ? N Feedback Pass ? Pass ? Annotator training Annotating images (2017, methodology): “State-of-the-art WSOL approaches based on AlexNet architecture [34] perform as follows.”
Weakly Supervised Object Detection with Posterior Regularization (2014, methodology): “While [5] represents these windows with the high-dimensional Fisher Vectors [22], [24] relies on the CNN features that are also used in this paper.”
Finding "It": Weakly-Supervised Reference-Aware Visual Grounding in Instructional Videos (2018, methodology): “MIL has been a effective framework for weakly-supervised learning in several applications, including image classification [50], object localization [13], tracking [5], and instance segmentation [37].”
Spot On: Action Localization from Pointly-Supervised Proposals (2016, methodology): “We draw inspiration from related work on weakly-supervised object detection, e.g., [6,7,8].”
Visual Attention Methods in Deep Learning: An In-Depth Survey (2022, background): “In weakly-supervised object localization, detecting the whole object without location annotation is a challenging task [136, 136, 44].”
Detector discovery in the wild: Joint multiple instance and representation learning (2014, methodology): “Approaches to better optimize the non-convex problem involve using multifold learning as a measure of regularizing overfitting [8], optimize Latent SVM for the area under the ROC curve (AUC) [6] and training with easy examples in beginning to avoid bad local optimization [5, 24].”
Efficient Image and Video Co-localization with Frank-Wolfe Algorithm (2014, unknown): “Method Our method (Co-localization) [44] (WSL) [46] (WSL) [45] (WSL) [43] (WSL) [17](WSL)”
Weakly Supervised Deep Detection Networks (2015, background): “te of the art in weakly supervised detection literature in Table 1 and 2. The results show that our method achieves overall better or comparable performance to these alternatives. While Cinbis et al. [4] use Fisher Vectors [28] along with SSW, the majority of previous work [32, 33, 1, 37, 2] use the”
Weakly-Supervised Video Object Grounding from Text by Loss Weighting and Object Interaction (2018, background): “Weakly-supervised object localization has been explored in both the image [2, 4, 5, 14, 21] and the video domain [10, 17].”
Bayesian Joint Modelling for Object Localisation in Weakly Labelled Images (2015, background): “Very recently, [68] achieved similar performance by training a multi-instance SVM with a more powerful fisher vector based representation.”