Search Results for author:
Found 94 papers, 53 papers with code
3D-Aware Multi-Class Image-to-Image Translation with NeRFs
In the second step, based on the well-trained multi-class 3D-aware GAN architecture, that preserves view-consistency, we construct a 3D-aware I2I translation system.
ICICLE: Interpretable Class Incremental Continual Learning
Continual learning enables incremental learning of new tasks without forgetting those previously learned, resulting in positive knowledge transfer that can enhance performance on both new and old tasks.
Towards Label-Efficient Incremental Learning: A Survey
The current dominant paradigm when building a machine learning model is to iterate over a dataset over and over until convergence.
Exemplar-free Continual Learning of Vision Transformers via Gated Class-Attention and Cascaded Feature Drift Compensation
Secondly, we propose a new method of feature drift compensation that accommodates feature drift in the backbone when learning new tasks.
Attribution-aware Weight Transfer: A Warm-Start Initialization for Class-Incremental Semantic Segmentation
In class-incremental semantic segmentation (CISS), deep learning architectures suffer from the critical problems of catastrophic forgetting and semantic background shift.
Ranked #1 on
Overlapped 14-1
on Cityscapes
Positive Pair Distillation Considered Harmful: Continual Meta Metric Learning for Lifelong Object Re-Identification
Lifelong object re-identification incrementally learns from a stream of re-identification tasks.
Attention Distillation: self-supervised vision transformer students need more guidance
In experiments on ImageNet-Subset and ImageNet-1K, we show that our method AttnDistill outperforms existing self-supervised knowledge distillation (SSKD) methods and achieves state-of-the-art k-NN accuracy compared with self-supervised learning (SSL) methods learning from scratch (with the ViT-S model).
OneRing: A Simple Method for Source-free Open-partial Domain Adaptation
In this paper, we investigate Source-free Open-partial Domain Adaptation (SF-OPDA), which addresses the situation where there exist both domain and category shifts between source and target domains.
MVMO: A Multi-Object Dataset for Wide Baseline Multi-View Semantic Segmentation
We present MVMO (Multi-View, Multi-Object dataset): a synthetic dataset of 116, 000 scenes containing randomly placed objects of 10 distinct classes and captured from 25 camera locations in the upper hemisphere.
Attracting and Dispersing: A Simple Approach for Source-free Domain Adaptation
Treating SFDA as an unsupervised clustering problem and following the intuition that local neighbors in feature space should have more similar predictions than other features, we propose to optimize an objective of prediction consistency.
Towards Exemplar-Free Continual Learning in Vision Transformers: an Account of Attention, Functional and Weight Regularization
In this paper, we investigate the continual learning of Vision Transformers (ViT) for the challenging exemplar-free scenario, with special focus on how to efficiently distill the knowledge of its crucial self-attention mechanism (SAM).
Planckian Jitter: countering the color-crippling effects of color jitter on self-supervised training
The data augmentations used are of crucial importance to the quality of learned feature representations.
Main Product Detection with Graph Networks for Fashion
Computer vision has established a foothold in the online fashion retail industry.
Continually Learning Self-Supervised Representations with Projected Functional Regularization
Recent self-supervised learning methods are able to learn high-quality image representations and are closing the gap with supervised approaches.
Visual Transformers with Primal Object Queries for Multi-Label Image Classification
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
+2
Transferring Unconditional to Conditional GANs with Hyper-Modulation
GANs have matured in recent years and are able to generate high-resolution, realistic images.
Incremental Meta-Learning via Episodic Replay Distillation for Few-Shot Image Recognition
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.
HCV: Hierarchy-Consistency Verification for Incremental Implicitly-Refined Classification
To overcome forgetting in this benchmark, we propose Hierarchy-Consistency Verification (HCV) as an enhancement to existing continual learning methods.
Class-Balanced Active Learning for Image Classification
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.
Exploiting the Intrinsic Neighborhood Structure for Source-free Domain Adaptation
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.
Distilling GANs with Style-Mixed Triplets for X2I Translation with Limited Data
Therefore, we investigate knowledge distillation to transfer knowledge from a high-quality unconditioned generative model (e. g., StyleGAN) to a conditioned synthetic image generation modules in a variety of systems.
Reducing Label Effort: Self-Supervised meets Active Learning
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.
Generalized Source-free Domain Adaptation
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.
When Deep Learners Change Their Mind: Learning Dynamics for Active Learning
Many methods approach this problem by measuring the informativeness of samples and do this based on the certainty of the network predictions for samples.
On the importance of cross-task features for class-incremental learning
We also define a new forgetting measure for class-incremental learning, and see that forgetting is not the principal cause of low performance.
ACAE-REMIND for Online Continual Learning with Compressed Feature Replay
Methods are typically allowed to use a limited buffer to store some of the images in the stream.
TransferI2I: Transfer Learning for Image-to-Image Translation from Small Datasets
In this paper, we propose a new transfer learning for I2I translation (TransferI2I).
MineGAN++: Mining Generative Models for Efficient Knowledge Transfer to Limited Data Domains
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.
Continual learning in cross-modal retrieval
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.
Avalanche: an End-to-End Library for Continual Learning
Learning continually from non-stationary data streams is a long-standing goal and a challenging problem in machine learning.
On Implicit Attribute Localization for Generalized Zero-Shot Learning
Zero-shot learning (ZSL) aims to discriminate images from unseen classes by exploiting relations to seen classes via their attribute-based descriptions.
Generative Multi-Label Zero-Shot Learning
Nevertheless, computing reliable attention maps for unseen classes during inference in a multi-label setting is still a challenge.
Ranked #8 on
Multi-label zero-shot learning
on NUS-WIDE
Self-Training for Class-Incremental Semantic Segmentation
In class-incremental semantic segmentation, we have no access to the labeled data of previous tasks.
Class-Incremental Semantic Segmentation
Incremental Learning
DeepI2I: Enabling Deep Hierarchical Image-to-Image Translation by Transferring from GANs
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.
Class-incremental learning: survey and performance evaluation on image classification
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.
Casting a BAIT for Offline and Online Source-free Domain Adaptation
When adapting to the target domain, the additional classifier initialized from source classifier is expected to find misclassified features.
Source-Free Domain Adaptation
Unsupervised Domain Adaptation
Learning to Rank for Active Learning: A Listwise Approach
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.).
Hallucinating Saliency Maps for Fine-Grained Image Classification for Limited Data Domains
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.
Disentanglement of Color and Shape Representations for Continual Learning
We hypothesize that disentangled feature representations suffer less from catastrophic forgetting.
RATT: Recurrent Attention to Transient Tasks for Continual Image Captioning
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.
On Class Orderings for Incremental Learning
The influence of class orderings in the evaluation of incremental learning has received very little attention.
Bookworm continual learning: beyond zero-shot learning and continual learning
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.
Simple and effective localized attribute representations for zero-shot learning
Zero-shot learning (ZSL) aims to discriminate images from unseen classes by exploiting relations to seen classes via their semantic descriptions.
Distributed Learning and Inference with Compressed Images
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.
Generative Feature Replay For Class-Incremental Learning
To prevent forgetting, we combine generative feature replay in the classifier with feature distillation in the feature extractor.
Semantic Drift Compensation for Class-Incremental Learning
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.
Semi-supervised Learning for Few-shot Image-to-Image Translation
In this work, we go one step further and reduce the amount of required labeled data also from the source domain during training.
Ternary Feature Masks: zero-forgetting for task-incremental learning
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.
Learning to adapt class-specific features across domains for semantic segmentation
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.
MineGAN: effective knowledge transfer from GANs to target domains with few images
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.
Variable Rate Deep Image Compression with Modulated Autoencoder
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.
Orderless Recurrent Models for Multi-label Classification
Recurrent neural networks (RNN) are popular for many computer vision tasks, including multi-label classification.
Active Learning for Deep Detection Neural Networks
In this paper, we propose a method to perform active learning of object detectors based on convolutional neural networks.
Multi-Modal Fusion for End-to-End RGB-T Tracking
Our tracker is trained in an end-to-end manner, enabling the components to learn how to fuse the information from both modalities.
Temporal Coherence for Active Learning in Videos
Our active learning criterion is based on the estimated number of errors in terms of false positives and false negatives.
Self-supervised blur detection from synthetically blurred scenes
Blur detection aims at segmenting the blurred areas of a given image.
SDIT: Scalable and Diverse Cross-domain Image Translation
Recently, image-to-image translation research has witnessed remarkable progress.
Learning the Model Update for Siamese Trackers
In general, this template is linearly combined with the accumulated template from the previous frame, resulting in an exponential decay of information over time.
Controlling biases and diversity in diverse image-to-image translation
The task of unpaired image-to-image translation is highly challenging due to the lack of explicit cross-domain pairs of instances.
Sparse data interpolation using the geodesic distance affinity space
In this paper, we adapt the geodesic distance-based recursive filter to the sparse data interpolation problem.
Learning Metrics from Teachers: Compact Networks for Image Embedding
In this paper, we propose to use network distillation to efficiently compute image embeddings with small networks.
Mix and match networks: cross-modal alignment for zero-pair image-to-image translation
This paper addresses the problem of inferring unseen cross-modal image-to-image translations between multiple modalities.
Exploiting Unlabeled Data in CNNs by Self-supervised Learning to Rank
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.
One-view occlusion detection for stereo matching with a fully connected CRF model
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.
Deep Spectral Reflectance and Illuminant Estimation from Self-Interreflections
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
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
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
This paper describes the multimodal Neural Machine Translation systems developed by LIUM and CVC for WMT18 Shared Task on Multimodal Translation.
Metric Learning for Novelty and Anomaly Detection
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.
Saliency for Fine-grained Object Recognition in Domains with Scarce Training Data
Our proposed pipeline allows to evaluate saliency methods for the high-level task of object recognition.
Context Proposals for Saliency Detection
Therefore, we extend object proposal methods with context proposals, which allow to incorporate the immediate context in the saliency computation.
Synthetic data generation for end-to-end thermal infrared tracking
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 for cross-domain disentanglement
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.
Learning Illuminant Estimation from Object Recognition
In this paper we present a deep learning method to estimate the illuminant of an image.
Weakly Supervised Domain-Specific Color Naming Based on Attention
The attention branch is used to modulate the pixel-wise color naming predictions of the network.
Transferring GANs: generating images from limited data
Transferring the knowledge of pretrained networks to new domains by means of finetuning is a widely used practice for applications based on discriminative models.
Mix and match networks: encoder-decoder alignment for zero-pair image translation
We address the problem of image translation between domains or modalities for which no direct paired data is available (i. e. zero-pair translation).
Leveraging Unlabeled Data for Crowd Counting by Learning to Rank
We propose a novel crowd counting approach that leverages abundantly available unlabeled crowd imagery in a learning-to-rank framework.
Ranked #17 on
Crowd Counting
on ShanghaiTech B
Rotate your Networks: Better Weight Consolidation and Less Catastrophic Forgetting
In this paper we propose an approach to avoiding catastrophic forgetting in sequential task learning scenarios.
Domain-adaptive deep network compression
We show that domain transfer leads to large shifts in network activations and that it is desirable to take this into account when compressing.
Adversarial Networks for Spatial Context-Aware Spectral Image Reconstruction from RGB
Hyperspectral signal reconstruction aims at recovering the original spectral input that produced a certain trichromatic (RGB) response from a capturing device or observer.
Review on Computer Vision Techniques in Emergency Situation
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
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.
LIUM-CVC Submissions for WMT17 Multimodal Translation Task
This paper describes the monomodal and multimodal Neural Machine Translation systems developed by LIUM and CVC for WMT17 Shared Task on Multimodal Translation.
Binary Patterns Encoded Convolutional Neural Networks for Texture Recognition and Remote Sensing Scene Classification
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.
Ranked #9 on
Aerial Scene Classification
on AID (20% as trainset)
Bandwidth limited object recognition in high resolution imagery
This paper proposes a novel method to optimize bandwidth usage for object detection in critical communication scenarios.
Scale Coding Bag of Deep Features for Human Attribute and Action Recognition
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.
Ensembles of Generative Adversarial Networks
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
Generative Adversarial Networks (GANs) have recently demonstrated to successfully approximate complex data distributions.
Ranked #4 on
Image-to-Image Translation
on RaFD
Does Multimodality Help Human and Machine for Translation and Image Captioning?
This paper presents the systems developed by LIUM and CVC for the WMT16 Multimodal Machine Translation challenge.
On-the-fly Network Pruning for Object Detection
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.
From Emotions to Action Units With Hidden and Semi-Hidden-Task Learning
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.
Adaptive Color Attributes for Real-Time Visual Tracking
This paper investigates the contribution of color in a tracking-by-detection framework.
Discriminative Color Descriptors
This results in a drop of discriminative power of the color description.
