Efficient Image and Video Co-localization with Frank-Wolfe Algorithm

By Armand Joulin, Kevin D. Tang, Li Fei-Fei. 2014

213 citations recorded by the source corpus.

Subfields: Aerospace Engineering, 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 Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

End-to-End Learning of Visual Representations From Uncurated Instructional Videos (2019, methodology): “The Frank-Wolfe (FW) algorithm has been shown effective for optimizing convex relaxation of (3) in a number of vision problems [1, 4, 5, 6, 19, 31].”
Joint Discovery of Object States and Manipulation Actions (2017, background): “Part of our object state model is related to [25], while our action model is related to [7].”
Minding the Gaps for Block Frank-Wolfe Optimization of Structured SVMs (2016, background): “The first three contributions are general to BCFW and thus could be applied to other block-separable optimization problems where BCFW could or have been used such as video colocalization (Joulin et al., 2014), multiple sequence alignment (Alayrac et al., 2016, App.”
Unsupervised Learning from Narrated Instruction Videos (2015, background): “In contrast, we show in the appendix of the paper [2] that the multiple sequence alignment problem given by (1) can be reformulated as an integer quadratic program with combinatorial constraints, for which the FrankWolfe optimization algorithm has been used recently with increasing success [5, 14, 1”
Grounding of Textual Phrases in Images by Reconstruction (2015, methodology): “[15] performs co-localization on videos using temporal consistency terms and turns the problem into integer program using Frank-Wolfe algorithm.”
Weakly-Supervised Alignment of Video with Text (2015, methodology): “Contrary to standard clustering techniques, it uses a discriminative cost function [2, 16] and it has been used in image co-segmentation [20, 21], object colocalization [22], person identification in video [6, 36], and alignment of labels to videos [7].”
Weakly-Supervised Learning of Visual Relations (2017, methodology): “Our work uses a discriminative clustering objective [3], that has been successful in several computer vision tasks [5, 24], but has not been so far, to the best of our knowledge, used for modeling relations.”
Unsupervised Object Discovery and Tracking in Video Collections (2015, methodology): “Following [17], our experiments are conducted on all the annotated videos.”
Unsupervised object discovery and localization in the wild: Part-based matching with bottom-up region proposals (2015, methodology): “The degree of supervision used in visual learning tasks varies from strong (supervised localization [13, 15]) to weak (weakly-supervised localization [8, 38]), very weak (colocalization [21, 40] and cosegmentation [33]), and null (fully-unsupervised discovery).”
Discover and Learn New Objects from Documentaries (2017, methodology): “[16] propose a method to localize objects of the same class across a set of distinct images or videos.”
Object Detection from Video Tubelets with Convolutional Neural Networks (2016, methodology): “Object localization and co-localization [27, 25, 14, 18], which have mainly focused on the YouTube Object Dataset (YTO) [27], seems to be a similar topic to the VID task.”
Blended Conditional Gradients: the unconditioning of conditional gradients (2018, unknown): “More recently, linear convergence has been established for variouspairwise-stepand away-stepvariants of CG in Lacoste-Julien and Jaggi [2015], where the concept of an active vertex set is used to improve performance.”
Lazifying Conditional Gradient Algorithms (2016, background): “In (Joulin et al., 2014) it is shown that video colocalization can be reduced to optimizing a quadratic objective function over a flow or a path polytope , which is the problem we are going to solve.”
Frank-Wolfe with Subsampling Oracle (2018, background): “…to their low memory requirements and projection-free iterations, which make them particularly appropriate to solve large scale convex problems, for instance convex relaxations of problems written over combinatorial polytopes (Zaslavskiy et al., 2009; Joulin et al., 2014; Vogelstein et al., 2015).”
Boosting Variational Inference: an Optimization Perspective (2017, background): “The algorithm is well studied both theoretically and empirically [14, 9, 8], and has even been applied to non-euclidean spaces.”
Boosting Frank-Wolfe by Chasing Gradients (2020, methodology): “…to traffic assignment problems (LeBlanc et al., 1975), low-rank matrix approximation (Shalev-Shwartz et al., 2011), structural SVMs (Lacoste-Julien et al., 2013), video co-localization (Joulin et al., 2014), infinite RBMs (Ping et al., 2016), and, e.g., adversarial learning (Chen et al., 2020).”