108 citations · 587 across the 36 of their papers we have counts for
8 papers · 1 filter
Full Contextual Attention for Multi-resolution Transformers in Semantic Segmentation
Loic Themyr, Clement Rambour, Nicolas Thome +2
Transformers have proved to be very effective for visual recognition tasks. In particular, vision transformers construct compressed global representations through self-attention an…
Vision and Structured-Language Pretraining for Cross-Modal Food Retrieval
Mustafa Shukor, Nicolas Thome, Matthieu Cord
Vision-Language Pretraining (VLP) and Foundation models have been the go-to recipe for achieving SoTA performance on general benchmarks. However, leveraging these powerful techniqu…
Memory transformers for full context and high-resolution 3D Medical Segmentation
Loic Themyr, Clément Rambour, Nicolas Thome +2
Transformer models achieve state-of-the-art results for image segmentation. However, achieving long-range attention, necessary to capture global context, with high-resolution 3D im…
Take One Gram of Neural Features, Get Enhanced Group Robustness
Simon Roburin, Charles Corbière, Gilles Puy +4
Predictive performance of machine learning models trained with empirical risk minimization (ERM) can degrade considerably under distribution shifts. The presence of spurious correl…
Hierarchical Average Precision Training for Pertinent Image Retrieval
Elias Ramzi, Nicolas Audebert, Nicolas Thome +2
Image Retrieval is commonly evaluated with Average Precision (AP) or Recall@k. Yet, those metrics, are limited to binary labels and do not take into account errors' severity. This…
Complementing Brightness Constancy with Deep Networks for Optical Flow Prediction
Vincent Le Guen, Clément Rambour, Nicolas Thome
State-of-the-art methods for optical flow estimation rely on deep learning, which require complex sequential training schemes to reach optimal performances on real-world data. In t…
Swapping Semantic Contents for Mixing Images
Rémy Sun, Clément Masson, Gilles Hénaff +2
Deep architecture have proven capable of solving many tasks provided a sufficient amount of labeled data. In fact, the amount of available labeled data has become the principal bot…
Towards efficient feature sharing in MIMO architectures
Rémy Sun, Alexandre Ramé, Clément Masson +2
Multi-input multi-output architectures propose to train multiple subnetworks within one base network and then average the subnetwork predictions to benefit from ensembling for free…