Search Results for author: Joost Van de Weijer

Found 80 papers, 39 papers with code

Visual Transformers with Primal Object Queries for Multi-Label Image Classification

no code implementations10 Dec 2021 Vacit Oguz Yazici, Joost Van de Weijer, LongLong Yu

However, inputting the same set of object queries to different decoder layers hinders the training: it results in lower performance and delays convergence.

Multi-Label Classification Multi-Label Image Classification +1

Hyper-GAN: Transferring Unconditional to Conditional GANs with HyperNetworks

no code implementations4 Dec 2021 Héctor Laria, Yaxing Wang, Joost Van de Weijer, Bogdan Raducanu

In extensive experiments, we validate the efficiency of the hypernetworks, self-initialization and contrastive loss for knowledge transfer on several standard benchmarks.

Transfer Learning

Incremental Meta-Learning via Episodic Replay Distillation for Few-Shot Image Recognition

no code implementations9 Nov 2021 Kai Wang, Xialei Liu, Andy Bagdanov, Luis Herranz, Shangling Jui, Joost Van de Weijer

We propose an approach to IML, which we call Episodic Replay Distillation (ERD), that mixes classes from the current task with class exemplars from previous tasks when sampling episodes for meta-learning.

Continual Learning Knowledge Distillation +1

HCV: Hierarchy-Consistency Verification for Incremental Implicitly-Refined Classification

1 code implementation21 Oct 2021 Kai Wang, Xialei Liu, Luis Herranz, Joost Van de Weijer

To overcome forgetting in this benchmark, we propose Hierarchy-Consistency Verification (HCV) as an enhancement to existing continual learning methods.

Continual Learning Incremental Learning

Class-Balanced Active Learning for Image Classification

1 code implementation9 Oct 2021 Javad Zolfaghari Bengar, Joost Van de Weijer, Laura Lopez Fuentes, Bogdan Raducanu

Results on three datasets showed that the method is general (it can be combined with most existing active learning algorithms) and can be effectively applied to boost the performance of both informative and representative-based active learning methods.

Active Learning Image Classification

Exploiting the Intrinsic Neighborhood Structure for Source-free Domain Adaptation

1 code implementation NeurIPS 2021 Shiqi Yang, Yaxing Wang, Joost Van de Weijer, Luis Herranz, Shangling Jui

In this paper, we address the challenging source-free domain adaptation (SFDA) problem, where the source pretrained model is adapted to the target domain in the absence of source data.

Domain Adaptation

Reducing Label Effort: Self-Supervised meets Active Learning

no code implementations25 Aug 2021 Javad Zolfaghari Bengar, Joost Van de Weijer, Bartlomiej Twardowski, Bogdan Raducanu

Our experiments reveal that self-training is remarkably more efficient than active learning at reducing the labeling effort, that for a low labeling budget, active learning offers no benefit to self-training, and finally that the combination of active learning and self-training is fruitful when the labeling budget is high.

Active Learning Object Recognition

Generalized Source-free Domain Adaptation

1 code implementation ICCV 2021 Shiqi Yang, Yaxing Wang, Joost Van de Weijer, Luis Herranz, Shangling Jui

In this paper, we propose a new domain adaptation paradigm called Generalized Source-free Domain Adaptation (G-SFDA), where the learned model needs to perform well on both the target and source domains, with only access to current unlabeled target data during adaptation.

Domain Adaptation

When Deep Learners Change Their Mind: Learning Dynamics for Active Learning

no code implementations30 Jul 2021 Javad Zolfaghari Bengar, Bogdan Raducanu, Joost Van de Weijer

Many methods approach this problem by measuring the informativeness of samples and do this based on the certainty of the network predictions for samples.

Active Learning Informativeness

On the importance of cross-task features for class-incremental learning

1 code implementation22 Jun 2021 Albin Soutif--Cormerais, Marc Masana, Joost Van de Weijer, Bartłomiej Twardowski

We also define a new forgetting measure for class-incremental learning, and see that forgetting is not the principal cause of low performance.

class-incremental learning Incremental Learning +1

ACAE-REMIND for Online Continual Learning with Compressed Feature Replay

no code implementations18 May 2021 Kai Wang, Luis Herranz, Joost Van de Weijer

Methods are typically allowed to use a limited buffer to store some of the images in the stream.

Continual Learning

MineGAN++: Mining Generative Models for Efficient Knowledge Transfer to Limited Data Domains

1 code implementation28 Apr 2021 Yaxing Wang, Abel Gonzalez-Garcia, Chenshen Wu, Luis Herranz, Fahad Shahbaz Khan, Shangling Jui, Joost Van de Weijer

Therefore, we propose a novel knowledge transfer method for generative models based on mining the knowledge that is most beneficial to a specific target domain, either from a single or multiple pretrained GANs.

Transfer Learning

Continual learning in cross-modal retrieval

no code implementations14 Apr 2021 Kai Wang, Luis Herranz, Joost Van de Weijer

We found that the indexing stage pays an important role and that simply avoiding reindexing the database with updated embedding networks can lead to significant gains.

Continual Learning Cross-Modal Retrieval

On Implicit Attribute Localization for Generalized Zero-Shot Learning

no code implementations8 Mar 2021 Shiqi Yang, Kai Wang, Luis Herranz, Joost Van de Weijer

Zero-shot learning (ZSL) aims to discriminate images from unseen classes by exploiting relations to seen classes via their attribute-based descriptions.

Generalized Zero-Shot Learning

Generative Multi-Label Zero-Shot Learning

1 code implementation27 Jan 2021 Akshita Gupta, Sanath Narayan, Salman Khan, Fahad Shahbaz Khan, Ling Shao, Joost Van de Weijer

Nevertheless, computing reliable attention maps for unseen classes during inference in a multi-label setting is still a challenge.

Image Classification Multi-label zero-shot learning

Self-Training for Class-Incremental Semantic Segmentation

no code implementations6 Dec 2020 Lu Yu, Xialei Liu, Joost Van de Weijer

In class-incremental semantic segmentation we have no access to the labeled data of previous tasks.

Incremental Learning Semantic Segmentation

DeepI2I: Enabling Deep Hierarchical Image-to-Image Translation by Transferring from GANs

1 code implementation NeurIPS 2020 Yaxing Wang, Lu Yu, Joost Van de Weijer

To enable the training of deep I2I models on small datasets, we propose a novel transfer learning method, that transfers knowledge from pre-trained GANs.

Image-to-Image Translation Transfer Learning +1

Class-incremental learning: survey and performance evaluation on image classification

1 code implementation28 Oct 2020 Marc Masana, Xialei Liu, Bartlomiej Twardowski, Mikel Menta, Andrew D. Bagdanov, Joost Van de Weijer

For future learning systems incremental learning is desirable, because it allows for: efficient resource usage by eliminating the need to retrain from scratch at the arrival of new data; reduced memory usage by preventing or limiting the amount of data required to be stored -- also important when privacy limitations are imposed; and learning that more closely resembles human learning.

class-incremental learning General Classification +2

Casting a BAIT for Offline and Online Source-free Domain Adaptation

1 code implementation23 Oct 2020 Shiqi Yang, Yaxing Wang, Joost Van de Weijer, Luis Herranz, Shangling Jui

When adapting to the target domain, the additional classifier initialized from source classifier is expected to find misclassified features.

Unsupervised Domain Adaptation

Learning to Rank for Active Learning: A Listwise Approach

no code implementations31 Jul 2020 Minghan Li, Xialei Liu, Joost Van de Weijer, Bogdan Raducanu

Active learning emerged as an alternative to alleviate the effort to label huge amount of data for data hungry applications (such as image/video indexing and retrieval, autonomous driving, etc.).

Active Learning Autonomous Driving +2

Hallucinating Saliency Maps for Fine-Grained Image Classification for Limited Data Domains

no code implementations24 Jul 2020 Carola Figueroa-Flores, Bogdan Raducanu, David Berga, Joost Van de Weijer

Most of the saliency methods are evaluated on their ability to generate saliency maps, and not on their functionality in a complete vision pipeline, like for instance, image classification.

Fine-Grained Image Classification General Classification +2

RATT: Recurrent Attention to Transient Tasks for Continual Image Captioning

1 code implementation NeurIPS 2020 Riccardo Del Chiaro, Bartłomiej Twardowski, Andrew D. Bagdanov, Joost Van de Weijer

We call our method Recurrent Attention to Transient Tasks (RATT), and also show how to adapt continual learning approaches based on weight egularization and knowledge distillation to recurrent continual learning problems.

Continual Learning Image Captioning +1

Disentanglement of Color and Shape Representations for Continual Learning

no code implementations13 Jul 2020 David Berga, Marc Masana, Joost Van de Weijer

We hypothesize that disentangled feature representations suffer less from catastrophic forgetting.

Continual Learning

On Class Orderings for Incremental Learning

no code implementations4 Jul 2020 Marc Masana, Bartłomiej Twardowski, Joost Van de Weijer

The influence of class orderings in the evaluation of incremental learning has received very little attention.

Incremental Learning

Bookworm continual learning: beyond zero-shot learning and continual learning

1 code implementation26 Jun 2020 Kai Wang, Luis Herranz, Anjan Dutta, Joost Van de Weijer

We propose bookworm continual learning(BCL), a flexible setting where unseen classes can be inferred via a semantic model, and the visual model can be updated continually.

Continual Learning Zero-Shot Learning

Simple and effective localized attribute representations for zero-shot learning

no code implementations10 Jun 2020 Shiqi Yang, Kai Wang, Luis Herranz, Joost Van de Weijer

Zero-shot learning (ZSL) aims to discriminate images from unseen classes by exploiting relations to seen classes via their semantic descriptions.

Zero-Shot Learning

Distributed Learning and Inference with Compressed Images

no code implementations22 Apr 2020 Sudeep Katakol, Basem Elbarashy, Luis Herranz, Joost Van de Weijer, Antonio M. Lopez

Moreover, we may only have compressed images at training time but are able to use original images at inference time, or vice versa, and in such a case, the downstream model suffers from covariate shift.

Autonomous Driving Image Compression +2

Semantic Drift Compensation for Class-Incremental Learning

1 code implementation CVPR 2020 Lu Yu, Bartłomiej Twardowski, Xialei Liu, Luis Herranz, Kai Wang, Yongmei Cheng, Shangling Jui, Joost Van de Weijer

The vast majority of methods have studied this scenario for classification networks, where for each new task the classification layer of the network must be augmented with additional weights to make room for the newly added classes.

class-incremental learning General Classification +1

Ternary Feature Masks: zero-forgetting for task-incremental learning

no code implementations23 Jan 2020 Marc Masana, Tinne Tuytelaars, Joost Van de Weijer

To allow already learned features to adapt to the current task without changing the behavior of these features for previous tasks, we introduce task-specific feature normalization.

Continual Learning Incremental Learning

Learning to adapt class-specific features across domains for semantic segmentation

1 code implementation22 Jan 2020 Mikel Menta, Adriana Romero, Joost Van de Weijer

Recent advances in unsupervised domain adaptation have shown the effectiveness of adversarial training to adapt features across domains, endowing neural networks with the capability of being tested on a target domain without requiring any training annotations in this domain.

Semantic Segmentation Translation +1

Variable Rate Deep Image Compression with Modulated Autoencoder

1 code implementation11 Dec 2019 Fei Yang, Luis Herranz, Joost Van de Weijer, José A. Iglesias Guitián, Antonio López, Mikhail Mozerov

Addressing these limitations, we formulate the problem of variable rate-distortion optimization for deep image compression, and propose modulated autoencoders (MAEs), where the representations of a shared autoencoder are adapted to the specific rate-distortion tradeoff via a modulation network.

Image Compression Video Compression

MineGAN: effective knowledge transfer from GANs to target domains with few images

2 code implementations CVPR 2020 Yaxing Wang, Abel Gonzalez-Garcia, David Berga, Luis Herranz, Fahad Shahbaz Khan, Joost Van de Weijer

We propose a novel knowledge transfer method for generative models based on mining the knowledge that is most beneficial to a specific target domain, either from a single or multiple pretrained GANs.

Transfer Learning

Multi-Modal Fusion for End-to-End RGB-T Tracking

1 code implementation30 Aug 2019 Lichao Zhang, Martin Danelljan, Abel Gonzalez-Garcia, Joost Van de Weijer, Fahad Shahbaz Khan

Our tracker is trained in an end-to-end manner, enabling the components to learn how to fuse the information from both modalities.

Image-to-Image Translation Rgb-T Tracking

Learning the Model Update for Siamese Trackers

1 code implementation ICCV 2019 Lichao Zhang, Abel Gonzalez-Garcia, Joost Van de Weijer, Martin Danelljan, Fahad Shahbaz Khan

In general, this template is linearly combined with the accumulated template from the previous frame, resulting in an exponential decay of information over time.

Visual Tracking

Controlling biases and diversity in diverse image-to-image translation

no code implementations23 Jul 2019 Yaxing Wang, Abel Gonzalez-Garcia, Joost Van de Weijer, Luis Herranz

The task of unpaired image-to-image translation is highly challenging due to the lack of explicit cross-domain pairs of instances.

Image-to-Image Translation Translation

Sparse data interpolation using the geodesic distance affinity space

no code implementations6 May 2019 Mikhail G. Mozerov, Fei Yang, Joost Van de Weijer

In this paper, we adapt the geodesic distance-based recursive filter to the sparse data interpolation problem.

Optical Flow Estimation

Exploiting Unlabeled Data in CNNs by Self-supervised Learning to Rank

2 code implementations17 Feb 2019 Xialei Liu, Joost Van de Weijer, Andrew D. Bagdanov

Our results show that networks trained to regress to the ground truth targets for labeled data and to simultaneously learn to rank unlabeled data obtain significantly better, state-of-the-art results for both IQA and crowd counting.

Active Learning Crowd Counting +4

One-view occlusion detection for stereo matching with a fully connected CRF model

no code implementations12 Jan 2019 Mikhail G. Mozerov, Joost Van de Weijer

We show that the OVOD approach considerably improves results for cost augmentation and energy minimization techniques in comparison with the standard one-view affinity space implementation.

Stereo Matching Stereo Matching Hand

Deep Spectral Reflectance and Illuminant Estimation from Self-Interreflections

no code implementations9 Dec 2018 Rada Deeb, Joost Van de Weijer, Damien Muselet, Mathieu Hebert, Alain Tremeau

In this work, we propose a CNN-based approach to estimate the spectral reflectance of a surface and the spectral power distribution of the light from a single RGB image of a V-shaped surface.

Memory Replay GANs: Learning to Generate New Categories without Forgetting

1 code implementation NeurIPS 2018 Chenshen Wu, Luis Herranz, Xialei Liu, Yaxing Wang, Joost Van de Weijer, Bogdan Raducanu

In particular, we investigate generative adversarial networks (GANs) in the task of learning new categories in a sequential fashion.

Memory Replay GANs: learning to generate images from new categories without forgetting

1 code implementation6 Sep 2018 Chenshen Wu, Luis Herranz, Xialei Liu, Yaxing Wang, Joost Van de Weijer, Bogdan Raducanu

In particular, we investigate generative adversarial networks (GANs) in the task of learning new categories in a sequential fashion.

LIUM-CVC Submissions for WMT18 Multimodal Translation Task

no code implementations WS 2018 Ozan Caglayan, Adrien Bardet, Fethi Bougares, Loïc Barrault, Kai Wang, Marc Masana, Luis Herranz, Joost Van de Weijer

This paper describes the multimodal Neural Machine Translation systems developed by LIUM and CVC for WMT18 Shared Task on Multimodal Translation.

Machine Translation Translation

Metric Learning for Novelty and Anomaly Detection

1 code implementation16 Aug 2018 Marc Masana, Idoia Ruiz, Joan Serrat, Joost Van de Weijer, Antonio M. Lopez

When neural networks process images which do not resemble the distribution seen during training, so called out-of-distribution images, they often make wrong predictions, and do so too confidently.

Anomaly Detection Metric Learning +2

Context Proposals for Saliency Detection

no code implementations27 Jun 2018 Aymen Azaza, Joost Van de Weijer, Ali Douik, Marc Masana

Therefore, we extend object proposal methods with context proposals, which allow to incorporate the immediate context in the saliency computation.

RGB Salient Object Detection Saliency Prediction +1

Synthetic data generation for end-to-end thermal infrared tracking

no code implementations4 Jun 2018 Lichao Zhang, Abel Gonzalez-Garcia, Joost Van de Weijer, Martin Danelljan, Fahad Shahbaz Khan

These methods provide us with a large labeled dataset of synthetic TIR sequences, on which we can train end-to-end optimal features for tracking.

Image-to-Image Translation Synthetic Data Generation +2

Image-to-image translation for cross-domain disentanglement

1 code implementation NeurIPS 2018 Abel Gonzalez-Garcia, Joost Van de Weijer, Yoshua Bengio

We compare our model to the state-of-the-art in multi-modal image translation and achieve better results for translation on challenging datasets as well as for cross-domain retrieval on realistic datasets.

Image-to-Image Translation Translation

Weakly Supervised Domain-Specific Color Naming Based on Attention

1 code implementation11 May 2018 Lu Yu, Yongmei Cheng, Joost Van de Weijer

The attention branch is used to modulate the pixel-wise color naming predictions of the network.

General Classification

Transferring GANs: generating images from limited data

1 code implementation ECCV 2018 Yaxing Wang, Chenshen Wu, Luis Herranz, Joost Van de Weijer, Abel Gonzalez-Garcia, Bogdan Raducanu

Transferring the knowledge of pretrained networks to new domains by means of finetuning is a widely used practice for applications based on discriminative models.

Domain Adaptation Image Generation

Mix and match networks: encoder-decoder alignment for zero-pair image translation

1 code implementation CVPR 2018 Yaxing Wang, Joost Van de Weijer, Luis Herranz

We address the problem of image translation between domains or modalities for which no direct paired data is available (i. e. zero-pair translation).

Colorization Semantic Segmentation +2

Leveraging Unlabeled Data for Crowd Counting by Learning to Rank

2 code implementations CVPR 2018 Xialei Liu, Joost Van de Weijer, Andrew D. Bagdanov

We propose a novel crowd counting approach that leverages abundantly available unlabeled crowd imagery in a learning-to-rank framework.

Crowd Counting Image Retrieval +1

Rotate your Networks: Better Weight Consolidation and Less Catastrophic Forgetting

2 code implementations8 Feb 2018 Xialei Liu, Marc Masana, Luis Herranz, Joost Van de Weijer, Antonio M. Lopez, Andrew D. Bagdanov

In this paper we propose an approach to avoiding catastrophic forgetting in sequential task learning scenarios.

Domain-adaptive deep network compression

2 code implementations ICCV 2017 Marc Masana, Joost Van de Weijer, Luis Herranz, Andrew D. Bagdanov, Jose M. Alvarez

We show that domain transfer leads to large shifts in network activations and that it is desirable to take this into account when compressing.

Low-rank compression

Adversarial Networks for Spatial Context-Aware Spectral Image Reconstruction from RGB

no code implementations1 Sep 2017 Aitor Alvarez-Gila, Joost Van de Weijer, Estibaliz Garrote

Hyperspectral signal reconstruction aims at recovering the original spectral input that produced a certain trichromatic (RGB) response from a capturing device or observer.

Image Reconstruction

Review on Computer Vision Techniques in Emergency Situation

no code implementations24 Aug 2017 Laura Lopez-Fuentes, Joost Van de Weijer, Manuel Gonzalez-Hidalgo, Harald Skinnemoen, Andrew D. Bagdanov

The number of emergencies where computer vision tools has been considered or used is very wide, and there is a great overlap across related emergency research.

RankIQA: Learning from Rankings for No-reference Image Quality Assessment

2 code implementations ICCV 2017 Xialei Liu, Joost Van de Weijer, Andrew D. Bagdanov

Furthermore, on the LIVE benchmark we show that our approach is superior to existing NR-IQA techniques and that we even outperform the state-of-the-art in full-reference IQA (FR-IQA) methods without having to resort to high-quality reference images to infer IQA.

No-Reference Image Quality Assessment

LIUM-CVC Submissions for WMT17 Multimodal Translation Task

no code implementations WS 2017 Ozan Caglayan, Walid Aransa, Adrien Bardet, Mercedes García-Martínez, Fethi Bougares, Loïc Barrault, Marc Masana, Luis Herranz, Joost Van de Weijer

This paper describes the monomodal and multimodal Neural Machine Translation systems developed by LIUM and CVC for WMT17 Shared Task on Multimodal Translation.

Machine Translation Translation

Binary Patterns Encoded Convolutional Neural Networks for Texture Recognition and Remote Sensing Scene Classification

no code implementations5 Jun 2017 Rao Muhammad Anwer, Fahad Shahbaz Khan, Joost Van de Weijer, Matthieu Molinier, Jorma Laaksonen

To the best of our knowledge, we are the first to investigate Binary Patterns encoded CNNs and different deep network fusion architectures for texture recognition and remote sensing scene classification.

Aerial Scene Classification General Classification +2

Bandwidth limited object recognition in high resolution imagery

no code implementations16 Jan 2017 Laura Lopez-Fuentes, Andrew D. Bagdanov, Joost Van de Weijer, Harald Skinnemoen

This paper proposes a novel method to optimize bandwidth usage for object detection in critical communication scenarios.

Information Seeking Object Detection +1

Scale Coding Bag of Deep Features for Human Attribute and Action Recognition

no code implementations14 Dec 2016 Fahad Shahbaz Khan, Joost Van de Weijer, Rao Muhammad Anwer, Andrew D. Bagdanov, Michael Felsberg, Jorma Laaksonen

Most approaches to human attribute and action recognition in still images are based on image representation in which multi-scale local features are pooled across scale into a single, scale-invariant encoding.

Action Recognition In Still Images

Ensembles of Generative Adversarial Networks

no code implementations3 Dec 2016 Yaxing Wang, Lichao Zhang, Joost Van de Weijer

The first one is based on the fact that in the minimax game which is played to optimize the GAN objective the generator network keeps on changing even after the network can be considered optimal.

Invertible Conditional GANs for image editing

6 code implementations19 Nov 2016 Guim Perarnau, Joost Van de Weijer, Bogdan Raducanu, Jose M. Álvarez

Generative Adversarial Networks (GANs) have recently demonstrated to successfully approximate complex data distributions.

Conditional Image Generation Image-to-Image Translation

On-the-fly Network Pruning for Object Detection

no code implementations11 May 2016 Marc Masana, Joost Van de Weijer, Andrew D. Bagdanov

Object detection with deep neural networks is often performed by passing a few thousand candidate bounding boxes through a deep neural network for each image.

Network Pruning Object Detection

From Emotions to Action Units With Hidden and Semi-Hidden-Task Learning

no code implementations ICCV 2015 Adria Ruiz, Joost Van de Weijer, Xavier Binefa

Additionally, we show that SHTL achieves competitive performance compared with state-of-the-art Transductive Learning approaches which face the problem of limited training data by using unlabelled test samples during training.

Cannot find the paper you are looking for? You can Submit a new open access paper.