2k citations · 6.6k across the 78 of their papers we have counts for
31 papers · 1 filter
DINOv3
Oriane Siméoni, Huy V. Vo, Maximilian Seitzer +23
Self-supervised learning holds the promise of eliminating the need for manual data annotation, enabling models to scale effortlessly to massive datasets and larger architectures. B…
Cluster and Predict Latent Patches for Improved Masked Image Modeling
Timothée Darcet, Federico Baldassarre, Maxime Oquab +2
Masked Image Modeling (MIM) offers a promising approach to self-supervised representation learning, however existing MIM models still lag behind the state-of-the-art. In this paper…
Fine Dense Alignment of Image Bursts through Camera Pose and Depth Estimation
Bruno Lecouat, Yann Dubois de Mont-Marin, Théo Bodrito +2
This paper introduces a novel approach to the fine alignment of images in a burst captured by a handheld camera. In contrast to traditional techniques that estimate two-dimensional…
Towards Real-World Focus Stacking with Deep Learning
Alexandre Araujo, Jean Ponce, Julien Mairal
Focus stacking is widely used in micro, macro, and landscape photography to reconstruct all-in-focus images from multiple frames obtained with focus bracketing, that is, with shall…
Vision Transformers Need Registers
Timothée Darcet, Maxime Oquab, Julien Mairal +1
Transformers have recently emerged as a powerful tool for learning visual representations. In this paper, we identify and characterize artifacts in feature maps of both supervised…
SUnAA: Sparse Unmixing using Archetypal Analysis
Behnood Rasti, Alexandre Zouaoui, Julien Mairal +1
This paper introduces a new sparse unmixing technique using archetypal analysis (SUnAA). First, we design a new model based on archetypal analysis. We assume that the endmembers of…
SLACK: Stable Learning of Augmentations with Cold-start and KL regularization
Juliette Marrie, Michael Arbel, Diane Larlus +1
Data augmentation is known to improve the generalization capabilities of neural networks, provided that the set of transformations is chosen with care, a selection often performed…
Semi-supervised learning made simple with self-supervised clustering
Enrico Fini, Pietro Astolfi, Karteek Alahari +4
Self-supervised learning models have been shown to learn rich visual representations without requiring human annotations. However, in many real-world scenarios, labels are partiall…
DINOv2: Learning Robust Visual Features without Supervision
Maxime Oquab, Timothée Darcet, Théo Moutakanni +23
The recent breakthroughs in natural language processing for model pretraining on large quantities of data have opened the way for similar foundation models in computer vision. Thes…
High Dynamic Range and Super-Resolution from Raw Image Bursts
Bruno Lecouat, Thomas Eboli, Jean Ponce +1
Photographs captured by smartphones and mid-range cameras have limited spatial resolution and dynamic range, with noisy response in underexposed regions and color artefacts in satu…
Self Supervised Learning for Few Shot Hyperspectral Image Classification
Nassim Ait Ali Braham, Lichao Mou, Jocelyn Chanussot +2
Deep learning has proven to be a very effective approach for Hyperspectral Image (HSI) classification. However, deep neural networks require large annotated datasets to generalize…
Self-Supervised Models are Continual Learners
Enrico Fini, Victor G. Turrisi da Costa, Xavier Alameda-Pineda +3
Self-supervised models have been shown to produce comparable or better visual representations than their supervised counterparts when trained offline on unlabeled data at scale. Ho…
NTIRE 2021 Challenge on Burst Super-Resolution: Methods and Results
Goutam Bhat, Martin Danelljan, Radu Timofte +25
This paper reviews the NTIRE2021 challenge on burst super-resolution. Given a RAW noisy burst as input, the task in the challenge was to generate a clean RGB image with 4 times hig…
Emerging Properties in Self-Supervised Vision Transformers
Mathilde Caron, Hugo Touvron, Ishan Misra +4
In this paper, we question if self-supervised learning provides new properties to Vision Transformer (ViT) that stand out compared to convolutional networks (convnets). Beyond the…
Lucas-Kanade Reloaded: End-to-End Super-Resolution from Raw Image Bursts
Bruno Lecouat, Jean Ponce, Julien Mairal
This presentation addresses the problem of reconstructing a high-resolution image from multiple lower-resolution snapshots captured from slightly different viewpoints in space and…
A Flexible Framework for Designing Trainable Priors with Adaptive Smoothing and Game Encoding
Bruno Lecouat, Jean Ponce, Julien Mairal
We introduce a general framework for designing and training neural network layers whose forward passes can be interpreted as solving non-smooth convex optimization problems, and wh…
Unsupervised Learning of Visual Features by Contrasting Cluster Assignments
Mathilde Caron, Ishan Misra, Julien Mairal +3
Unsupervised image representations have significantly reduced the gap with supervised pretraining, notably with the recent achievements of contrastive learning methods. These contr…
Selecting Relevant Features from a Multi-domain Representation for Few-shot Classification
Nikita Dvornik, Cordelia Schmid, Julien Mairal
Popular approaches for few-shot classification consist of first learning a generic data representation based on a large annotated dataset, before adapting the representation to new…
Pruning Convolutional Neural Networks with Self-Supervision
Mathilde Caron, Ari Morcos, Piotr Bojanowski +2
Convolutional neural networks trained without supervision come close to matching performance with supervised pre-training, but sometimes at the cost of an even higher number of par…
Fully Trainable and Interpretable Non-Local Sparse Models for Image Restoration
Bruno Lecouat, Jean Ponce, Julien Mairal
Non-local self-similarity and sparsity principles have proven to be powerful priors for natural image modeling. We propose a novel differentiable relaxation of joint sparsity that…
Unsupervised Pre-Training of Image Features on Non-Curated Data
Mathilde Caron, Piotr Bojanowski, Julien Mairal +1
Pre-training general-purpose visual features with convolutional neural networks without relying on annotations is a challenging and important task. Most recent efforts in unsupervi…
Diversity with Cooperation: Ensemble Methods for Few-Shot Classification
Nikita Dvornik, Cordelia Schmid, Julien Mairal
Few-shot classification consists of learning a predictive model that is able to effectively adapt to a new class, given only a few annotated samples. To solve this challenging prob…
On the Importance of Visual Context for Data Augmentation in Scene Understanding
Nikita Dvornik, Julien Mairal, Cordelia Schmid
Performing data augmentation for learning deep neural networks is known to be important for training visual recognition systems. By artificially increasing the number of training e…
Modeling Visual Context is Key to Augmenting Object Detection Datasets
Nikita Dvornik, Julien Mairal, Cordelia Schmid
Performing data augmentation for learning deep neural networks is well known to be important for training visual recognition systems. By artificially increasing the number of train…