Integrated and Detailed Image Understanding

The Curious Layperson: Fine-Grained Image Recognition without Expert Labels

Auteurs: Subhabrata Choudhury,Iro Laina,Christian Rupprecht,Andrea Vedaldi
Publié dans: 2021 British Machine Vision Conference, 2021
Éditeur: BMVA

Unsupervised Part Discovery from Contrastive Reconstruction

Auteurs: Subhabrata Choudhury, Iro Laina, Christian Rupprecht, Andrea Vedaldi
Publié dans: 2021 Neural Information Processing Systems (NeurIPS), 2021
Éditeur: MIT

Deep Image Prior

Auteurs: Victor Lempitsky, Andrea Vedaldi, Dmitry Ulyanov
Publié dans: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018, Page(s) 9446-9454, ISBN 978-1-5386-6420-9
Éditeur: IEEE
DOI: 10.1109/cvpr.2018.00984

Small Steps and Giant Leaps: Minimal Newton Solvers for Deep Learning

Auteurs: Joao Henriques, Sebastien Ehrhardt, Samuel Albanie, Andrea Vedaldi
Publié dans: 2019 IEEE/CVF International Conference on Computer Vision (ICCV), 2019, Page(s) 4762-4771, ISBN 978-1-7281-4803-8
Éditeur: IEEE
DOI: 10.1109/iccv.2019.00486

RELATE: Physically Plausible Multi-Object Scene Synthesis Using Structured Latent Spaces

Auteurs: Sebastien Ehrhardt, Oliver Groth, Aron Monszpart, Martin Engelcke, Ingmar Posner, Niloy Mitra, Andrea Vedaldi
Publié dans: 2020 Neural Information Processing Systems (NeurIPS), 2020
Éditeur: NeurIPS

There and Back Again: Revisiting Backpropagation Saliency Methods

Auteurs: Sylvestre-Alvise Rebuffi, Ruth Fong, Xu Ji, Andrea Vedaldi
Publié dans: 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020, Page(s) 8836-8845, ISBN 978-1-7281-7168-5
Éditeur: IEEE
DOI: 10.1109/cvpr42600.2020.00886

Modelling and unsupervised learning of symmetric deformable object categories

Auteurs: James Thewlis, Hakan Bilen, Andrea Vedaldi
Publié dans: 2018 Neural Information Processing Systems (NeurIPS), 2018
Éditeur: NeurIPS

A critical analysis of self-supervision, or what we can learn from a single image

Auteurs: Yuki Asano, Christian Rupprecht, Andrea Vedaldi
Publié dans: 2020 International Conference on Machine Learning (ICML), 2020
Éditeur: ICML

Quantifying Learnability and Describability of Visual Concepts Emerging in Representation Learning

Auteurs: Iro Laina, Ruth Fong, Andrea Vedaldi
Publié dans: 2020 Neural Information Processing Systems (NeurIPS), 2020
Éditeur: NeurIPS

Keeping Your Eye on the Ball: Trajectory Attention in Video Transformers

Auteurs: Mandela Patrick, Dylan Campbell, Yuki Asano, Ishan Misra, Florian Metze, Christoph Feichtenhofer, Andrea Vedaldi, Joao F. Henriques
Publié dans: 2021 Neural Information Processing Systems (NeurIPS), 2021
Éditeur: MIT

Labelling unlabelled videos from scratch with multi-modal self-supervision

Auteurs: Yuki Asano, Mandela Patrick, Christian Rupprecht, Andrea Vedaldi
Publié dans: 2020 Neural Information Processing Systems (NeurIPS), 2020
Éditeur: NeurIPS

Scrutinizing and De-Biasing Intuitive Physics with Neural Stethoscopes

Auteurs: Fabian Fuchs, Oliver Groth, Adam Kosiorek, Alex Bewley, Markus Wulfmeier, Andrea Vedaldi, Ingmar Posner
Publié dans: 2019 British Machine Vision Conference (BMVC), 2019
Éditeur: BMVA
DOI: 10.5244/c.33.114

Unsupervised Learning of Landmarks by Descriptor Vector Exchange

Auteurs: James Thewlis, Samuel Albanie, Hakan Bilen, Andrea Vedaldi
Publié dans: 2019 IEEE/CVF International Conference on Computer Vision (ICCV), 2019, Page(s) 6360-6370, ISBN 978-1-7281-4803-8
Éditeur: IEEE
DOI: 10.1109/iccv.2019.00646

Weakly Supervised Deep Detection Networks

Auteurs: Hakan Bilen, Andrea Vedaldi
Publié dans: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016, Page(s) 2846-2854, ISBN 978-1-4673-8851-1
Éditeur: IEEE
DOI: 10.1109/CVPR.2016.311

Dynamic Image Networks for Action Recognition

Auteurs: Hakan Bilen, Basura Fernando, Efstratios Gavves, Andrea Vedaldi, Stephen Gould
Publié dans: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016, Page(s) 3034-3042, ISBN 978-1-4673-8851-1
Éditeur: IEEE
DOI: 10.1109/CVPR.2016.331

HPatches: A Benchmark and Evaluation of Handcrafted and Learned Local Descriptors

Auteurs: Vassileios Balntas, Karel Lenc, Andrea Vedaldi, Krystian Mikolajczyk
Publié dans: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Numéro 22-26 July 2017, 2017, Page(s) 3852-3861, ISBN 978-1-5386-0457-1
Éditeur: IEEE
DOI: 10.1109/CVPR.2017.410

Interpretable Explanations of Black Boxes by Meaningful Perturbation

Auteurs: Ruth C. Fong, Andrea Vedaldi
Publié dans: 2017 IEEE International Conference on Computer Vision (ICCV), Numéro 22-29 October 2017, 2017, Page(s) 3449-3457, ISBN 978-1-5386-1032-9
Éditeur: IEEE
DOI: 10.1109/ICCV.2017.371

Warped convolutions: efficient invariance to spatial transformation

Auteurs: Joao F. Henriques, Andrea Vedaldi
Publié dans: 2017 International Conference on Machine Learning (ICML), Numéro 7-9 August, 2017, 2017
Éditeur: ICML

AnchorNet: A Weakly Supervised Network to Learn Geometry-Sensitive Features for Semantic Matching

Auteurs: David Novotny, Diane Larlus, Andrea Vedaldi
Publié dans: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Numéro 22-26 July 2017, 2017, Page(s) 2867-2876, ISBN 978-1-5386-0457-1
Éditeur: IEEE
DOI: 10.1109/CVPR.2017.306

End-to-End Representation Learning for Correlation Filter Based Tracking

Auteurs: Jack Valmadre, Luca Bertinetto, Joao Henriques, Andrea Vedaldi, Philip H. S. Torr
Publié dans: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Numéro 22-26 July 2017, 2017, Page(s) 5000-5008, ISBN 978-1-5386-0457-1
Éditeur: IEEE
DOI: 10.1109/CVPR.2017.531

Learning Grimaces by Watching TV

Auteurs: Samuel Albanie, Andrea Vedaldi
Publié dans: Procedings of the British Machine Vision Conference 2016, Numéro 19-22 September, 2016, 2016, Page(s) 122.1-122.12, ISBN 1-901725-59-6
Éditeur: British Machine Vision Association
DOI: 10.5244/C.30.122

Fully-Convolutional Siamese Networks for Object Tracking

Auteurs: Luca Bertinetto, Jack Valmadre, João F. Henriques, Andrea Vedaldi, Philip H. S. Torr
Publié dans: ECCV 2016, Numéro 8-16 October 2016, 2016, Page(s) 850-865
Éditeur: Springer International Publishing
DOI: 10.1007/978-3-319-48881-3_56

Learning Covariant Feature Detectors

Auteurs: Karel Lenc, Andrea Vedaldi
Publié dans: ECCV Workshop on Geometry Meets Deep Learning, Numéro 9 October, 2016, 2016, Page(s) 100-117
Éditeur: Springer International Publishing
DOI: 10.1007/978-3-319-49409-8_11

Salient Deconvolutional Networks

Auteurs: Aravindh Mahendran, Andrea Vedaldi
Publié dans: ECCV, Numéro 8-16 October 2016, 2016, Page(s) 120-135
Éditeur: Springer International Publishing
DOI: 10.1007/978-3-319-46466-4_8

I Have Seen Enough: Transferring Parts Across Categories

Auteurs: David Novotny, Diane Larlus, Andrea Vedaldi
Publié dans: Procedings of the British Machine Vision Conference 2016, Numéro 19-22 September, 2016, Page(s) 115.1-115.14, ISBN 1-901725-59-6
Éditeur: British Machine Vision Association
DOI: 10.5244/C.30.115

Learning the Structure of Objects from Web Supervision

Auteurs: David Novotny, Diane Larlus, Andrea Vedaldi
Publié dans: Computer Vision – ECCV 2016 Workshops - Amsterdam, The Netherlands, October 8-10 and 15-16, 2016, Proceedings, Part III, Numéro 9915, 2016, Page(s) 218-235, ISBN 978-3-319-49408-1
Éditeur: Springer International Publishing
DOI: 10.1007/978-3-319-49409-8_19

Fully-trainable deep matching

Auteurs: James Thewlis, Shuai Zheng, Philip Torr, Andrea Vedaldi
Publié dans: Procedings of the British Machine Vision Conference 2016, Numéro 19-22 September, 2016, Page(s) 145.1-145.12, ISBN 1-901725-59-6
Éditeur: British Machine Vision Association
DOI: 10.5244/C.30.145

Learning feed-forward one-shot learners

Auteurs: Luca Bertinetto, Joao Henriques, Jack Valmadre, Philip Torr and Andrea Vedaldi
Publié dans: 2016 Neural Information Processing Systems (NIPS), 2016
Éditeur: Currant Associates, Inc.

Learning multiple visual domains with residual adapters

Auteurs: Sylvestre-Alvise Rebuffi, Hakan Bilen, Andrea Vedaldi
Publié dans: 2017 Neural Information Processing Systems (NeurIPS), 2017
Éditeur: Curran Associates, Inc.

Unsupervised learning of object frames by dense equivariant image labelling

Auteurs: James Thewlis, Hakan Bilen, Andrea Vedaldi
Publié dans: 2017 Neural Information Processing Systems (NeurIPS), 2017
Éditeur: Curran Associates, Inc.

Learning 3D Object Categories by Looking Around Them

Auteurs: David Novotny, Diane Larlus, Andrea Vedaldi
Publié dans: 2017 IEEE International Conference on Computer Vision (ICCV), 2017, Page(s) 5228-5237, ISBN 978-1-5386-1032-9
Éditeur: IEEE
DOI: 10.1109/iccv.2017.558

Integrated Perception with Recurrent Multi-Task Neural Networks

Auteurs: Hakan Bilen, Andrea Vedaldi
Publié dans: 2016 Neural Information Processing Systems (NIPS), 2016
Éditeur: Curran Associates, Inc.

Unsupervised Learning of Object Landmarks by Factorized Spatial Embeddings

Auteurs: James Thewlis, Hakan Bilen, Andrea Vedaldi
Publié dans: 2017 IEEE International Conference on Computer Vision (ICCV), 2017, Page(s) 3229-3238, ISBN 978-1-5386-1032-9
Éditeur: IEEE
DOI: 10.1109/iccv.2017.348

Efficient Parametrization of Multi-domain Deep Neural Networks

Auteurs: Sylvestre-Alvise Rebuffi, Andrea Vedaldi, Hakan Bilen
Publié dans: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018, Page(s) 8119-8127, ISBN 978-1-5386-6420-9
Éditeur: IEEE
DOI: 10.1109/cvpr.2018.00847

MapNet: An Allocentric Spatial Memory for Mapping Environments

Auteurs: Joao F. Henriques, Andrea Vedaldi
Publié dans: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018, Page(s) 8476-8484, ISBN 978-1-5386-6420-9
Éditeur: IEEE
DOI: 10.1109/cvpr.2018.00884

Gather-Excite: Exploiting Feature Context in Convolutional Neural Networks

Auteurs: Jie Hu, Li Shen, Samuel Albanie, Gang Sun, Andrea Vedaldi
Publié dans: 2018 Neural Information Processing Systems, 2018
Éditeur: NeurIPS

Unsupervised Learning of Object Landmarks through Conditional Image Generation

Auteurs: Tomas Jakab, Ankush Gupta, Hakan Bilen, Andrea Vedaldi
Publié dans: 2018 Neural Information Processing Systems (NeurIPS), 2018
Éditeur: NeurIPS

Large scale evaluation of local image feature detectors on homography datasets

Auteurs: Karel Lenc, Andrea Vedaldi
Publié dans: 2018 British Machine Vision Conference, 2018
Éditeur: BMVA

Cross Pixel Optical-Flow Similarity for Self-supervised Learning

Auteurs: Aravindh Mahendran, James Thewlis, Andrea Vedaldi
Publié dans: Computer Vision – ACCV 2018 - 14th Asian Conference on Computer Vision, Perth, Australia, December 2–6, 2018, Revised Selected Papers, Part V, Numéro 11365, 2019, Page(s) 99-116, ISBN 978-3-030-20872-1
Éditeur: Springer International Publishing
DOI: 10.1007/978-3-030-20873-8_7

Meta-learning with differentiable closed-form solvers

Auteurs: Luca Bertinetto, Joao F. Henriques, Philip Torr, Andrea Vedaldi
Publié dans: 2019 International Conference on Learning Representations (ICLR), 2019
Éditeur: ICLR

Learning to Discover Novel Visual Categories via Deep Transfer Clustering

Auteurs: Kai Han, Andrea Vedaldi, Andrew Zisserman
Publié dans: 2019 IEEE/CVF International Conference on Computer Vision (ICCV), 2019, Page(s) 8400-8408, ISBN 978-1-7281-4803-8
Éditeur: IEEE
DOI: 10.1109/iccv.2019.00849

Net2Vec: Quantifying and Explaining How Concepts are Encoded by Filters in Deep Neural Networks

Auteurs: Ruth Fong, Andrea Vedaldi
Publié dans: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018, Page(s) 8730-8738, ISBN 978-1-5386-6420-9
Éditeur: IEEE
DOI: 10.1109/cvpr.2018.00910

Unsupervised Intuitive Physics from Visual Observations

Auteurs: Sebastien Ehrhardt, Aron Monszpart, Niloy Mitra, Andrea Vedaldi
Publié dans: Computer Vision – ACCV 2018 - 14th Asian Conference on Computer Vision, Perth, Australia, December 2–6, 2018, Revised Selected Papers, Part III, Numéro 11363, 2019, Page(s) 700-716, ISBN 978-3-030-20892-9
Éditeur: Springer International Publishing
DOI: 10.1007/978-3-030-20893-6_44

Invariant Information Clustering for Unsupervised Image Classification and Segmentation

Auteurs: Xu Ji, Andrea Vedaldi, Joao Henriques
Publié dans: 2019 IEEE/CVF International Conference on Computer Vision (ICCV), 2019, Page(s) 9864-9873, ISBN 978-1-7281-4803-8
Éditeur: IEEE
DOI: 10.1109/iccv.2019.00996

Self-labelling via simultaneous clustering and representation learning

Auteurs: Yuki Asano, Christian Rupprecht, Andrea Vedaldi
Publié dans: 2020 International Conference on Learning Representations (ICLR), 2020
Éditeur: ICLR

Automatically Discovering and Learning New Visual Categories with Ranking Statistics

Auteurs: Kai Han, Sylvestre-Alvise Rebuffi, Sebastien Ehrhardt, Andrea Vedaldi, Andrew Zisserman
Publié dans: 2020 International Conference on Learning Representations (ICLR), 2020
Éditeur: ICLR

Self-Supervised Learning of Interpretable Keypoints From Unlabelled Videos

Auteurs: Tomas Jakab, Ankush Gupta, Hakan Bilen, Andrea Vedaldi
Publié dans: 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020, Page(s) 8784-8794, ISBN 978-1-7281-7168-5
Éditeur: IEEE
DOI: 10.1109/cvpr42600.2020.00881

Unsupervised Learning of Probably Symmetric Deformable 3D Objects From Images in the Wild

Auteurs: Shangzhe Wu, Christian Rupprecht, Andrea Vedaldi
Publié dans: 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020, Page(s) 1-10, ISBN 978-1-7281-7168-5
Éditeur: IEEE
DOI: 10.1109/cvpr42600.2020.00008

ShapeStacks: Learning Vision-Based Physical Intuition for Generalised Object Stacking

Auteurs: Oliver Groth, Fabian B. Fuchs, Ingmar Posner, Andrea Vedaldi
Publié dans: Computer Vision – ECCV 2018 - 15th European Conference, Munich, Germany, September 8-14, 2018, Proceedings, Part I, Numéro 11205, 2018, Page(s) 724-739, ISBN 978-3-030-01245-8
Éditeur: Springer International Publishing
DOI: 10.1007/978-3-030-01246-5_43

Self-Supervised Learning of Geometrically Stable Features Through Probabilistic Introspection

Auteurs: David Novotny, Samuel Albanie, Diane Larlus, Andrea Vedaldi
Publié dans: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018, Page(s) 3637-3645, ISBN 978-1-5386-6420-9
Éditeur: IEEE
DOI: 10.1109/cvpr.2018.00383

PASS: An ImageNet replacement for self-supervised pretraining without humans

Auteurs: Yuki M. Asano, Christian Rupprecht, Andrew Zisserman, Andrea Vedaldi
Publié dans: Proc. of the Intl. Conf. on Neural Information Processing Systems (NeurIPS) Track on Datasets and Benchmarks, 2021
Éditeur: NeurIPS foundation

It Takes (Only) Two: Adversarial Generator-Encoder Networks

Auteurs: Dmitry Ulyanov, Andrea Vedaldi, Victor Lempitsky
Publié dans: 2018 AAAI Conference on Artificial Intelligence, 2018
Éditeur: AAAI

AutoNovel: Automatically Discovering and Learning Novel Visual Categories

Auteurs: Kai Han, Sylvestre-Alvise Rebuffi, Sebastien Ehrhardt, Andrea Vedaldi, Andrew Zisserman
Publié dans: IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021, ISSN 0162-8828
Éditeur: Institute of Electrical and Electronics Engineers
DOI: 10.1109/tpami.2021.3091944

Deep Image Prior

Auteurs: Dmitry Ulyanov, Andrea Vedaldi, Victor Lempitsky
Publié dans: International Journal of Computer Vision, Numéro 128/7, 2020, Page(s) 1867-1888, ISSN 0920-5691
Éditeur: Kluwer Academic Publishers
DOI: 10.1007/s11263-020-01303-4

Action Recognition with Dynamic Image Networks

Auteurs: Hakan Bilen, Basura Fernando, Efstratios Gavves, Andrea Vedaldi
Publié dans: IEEE Transactions on Pattern Analysis and Machine Intelligence, Numéro 40/12, 2018, Page(s) 2799-2813, ISSN 0162-8828
Éditeur: Institute of Electrical and Electronics Engineers
DOI: 10.1109/tpami.2017.2769085

Visualizing Deep Convolutional Neural Networks Using Natural Pre-images

Auteurs: Aravindh Mahendran, Andrea Vedaldi
Publié dans: International Journal of Computer Vision, 2016, ISSN 0920-5691
Éditeur: Kluwer Academic Publishers
DOI: 10.1007/s11263-016-0911-8

Deep Filter Banks for Texture Recognition, Description, and Segmentation

Auteurs: Mircea Cimpoi, Subhransu Maji, Iasonas Kokkinos, Andrea Vedaldi
Publié dans: International Journal of Computer Vision, Numéro 118/1, 2016, Page(s) 65-94, ISSN 0920-5691
Éditeur: Kluwer Academic Publishers
DOI: 10.1007/s11263-015-0872-3

Understanding Image Representations by Measuring Their Equivariance and Equivalence

Auteurs: Karel Lenc, Andrea Vedaldi
Publié dans: International Journal of Computer Vision, 2018, ISSN 0920-5691
Éditeur: Kluwer Academic Publishers
DOI: 10.1007/s11263-018-1098-y

Taking visual motion prediction to new heightfields

Auteurs: Sebastien Ehrhardt, Aron Monszpart, Niloy J. Mitra, Andrea Vedaldi
Publié dans: Computer Vision and Image Understanding, Numéro 181, 2019, Page(s) 14-25, ISSN 1077-3142
Éditeur: Academic Press
DOI: 10.1016/j.cviu.2019.02.005

HPatches: A benchmark and evaluation of handcrafted and learned local descriptors

Auteurs: Vassileios Balntas, Karel Lenc, Andrea Vedaldi, Tinne Tuytelaars, Jiri Matas, Krystian Mikolajczyk
Publié dans: IEEE Transactions on Pattern Analysis and Machine Intelligence, 2019, Page(s) 1-1, ISSN 0162-8828
Éditeur: Institute of Electrical and Electronics Engineers
DOI: 10.1109/tpami.2019.2915233

Capturing the Geometry of Object Categories from Video Supervision

Auteurs: David Novotny, Diane Larlus, Andrea Vedaldi
Publié dans: IEEE Transactions on Pattern Analysis and Machine Intelligence, Numéro 42/2, 2020, Page(s) 261-275, ISSN 0162-8828
Éditeur: Institute of Electrical and Electronics Engineers
DOI: 10.1109/tpami.2018.2871117

Semi-convolutional Operators for Instance Segmentation

Auteurs: David Novotny, Samuel Albanie, Diane Larlus, Andrea Vedaldi
Publié dans: Computer Vision – ECCV 2018 - 15th European Conference, Munich, Germany, September 8-14, 2018, Proceedings, Part I, Numéro 11205, 2018, Page(s) 89-105, ISBN 978-3-030-01245-8
Éditeur: Springer International Publishing
DOI: 10.1007/978-3-030-01246-5_6

Publications

Partager cette page

Télécharger