Bruno Lecouat
YOU?
Author Swipe
View article: Fine Dense Alignment of Image Bursts through Camera Pose and Depth Estimation
Fine Dense Alignment of Image Bursts through Camera Pose and Depth Estimation Open
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 transformations between frame pairs or rely on discrete co…
View article: High Dynamic Range and Super-Resolution from Raw Image Bursts
High Dynamic Range and Super-Resolution from Raw Image Bursts Open
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 saturated areas. This paper introduces the first approach (to t…
View article: High dynamic range and super-resolution from raw image bursts
High dynamic range and super-resolution from raw image bursts Open
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 saturated areas. This paper introduces the first approach (to t…
View article: Lucas-Kanade Reloaded: End-to-End Super-Resolution from Raw Image Bursts
Lucas-Kanade Reloaded: End-to-End Super-Resolution from Raw Image Bursts Open
This presentation addresses the problem of reconstructing a high-resolution image from multiple lower-resolution snapshots captured from slightly different viewpoints in space and time. Key challenges for solving this problem include (i) a…
View article: NTIRE 2021 Challenge on Burst Super-Resolution: Methods and Results
NTIRE 2021 Challenge on Burst Super-Resolution: Methods and Results Open
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 higher resolution. The challenge contained two tracks; Track 1…
View article: Aliasing is your Ally: End-to-End Super-Resolution from Raw Image Bursts.
Aliasing is your Ally: End-to-End Super-Resolution from Raw Image Bursts. Open
This presentation addresses the problem of reconstructing a high-resolution image from multiple lower-resolution snapshots captured from slightly different viewpoints in space and time. Key challenges for solving this problem include (i) a…
View article: A Flexible Framework for Designing Trainable Priors with Adaptive Smoothing and Game Encoding
A Flexible Framework for Designing Trainable Priors with Adaptive Smoothing and Game Encoding Open
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 whose architectures are derived from an optimization algorith…
View article: Designing and Learning Trainable Priors with Non-Cooperative Games
Designing and Learning Trainable Priors with Non-Cooperative Games Open
We introduce a general framework for designing and learning neural networks whose forward passes can be interpreted as solving convex optimization problems, and whose architectures are derived from an optimization algorithm. We focus on no…
View article: A Flexible Framework for Designing Trainable Priors with Adaptive\n Smoothing and Game Encoding
A Flexible Framework for Designing Trainable Priors with Adaptive\n Smoothing and Game Encoding Open
We introduce a general framework for designing and training neural network\nlayers whose forward passes can be interpreted as solving non-smooth convex\noptimization problems, and whose architectures are derived from an optimization\nalgor…
View article: Fully Trainable and Interpretable Non-local Sparse Models for Image Restoration
Fully Trainable and Interpretable Non-local Sparse Models for Image Restoration Open
View article: Revisiting Non Local Sparse Models for Image Restoration
Revisiting Non Local Sparse Models for Image Restoration Open
We propose a differentiable algorithm for image restoration inspired by the success of sparse models and self-similarity priors for natural images. Our approach builds upon the concept of joint sparsity between groups of similar image patc…
View article: Fully Trainable and Interpretable Non-Local Sparse Models for Image Restoration
Fully Trainable and Interpretable Non-Local Sparse Models for Image Restoration Open
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 exploits both principles and leads to a general framework f…
View article: Venn GAN: Discovering Commonalities and Particularities of Multiple Distributions
Venn GAN: Discovering Commonalities and Particularities of Multiple Distributions Open
We propose a GAN design which models multiple distributions effectively and discovers their commonalities and particularities. Each data distribution is modeled with a mixture of $K$ generator distributions. As the generators are partially…
View article: Semi-Supervised Deep Learning for Abnormality Classification in Retinal Images
Semi-Supervised Deep Learning for Abnormality Classification in Retinal Images Open
Supervised deep learning algorithms have enabled significant performance gains in medical image classification tasks. But these methods rely on large labeled datasets that require resource-intensive expert annotation. Semi-supervised gener…
View article: Adversarially Learned Anomaly Detection
Adversarially Learned Anomaly Detection Open
Anomaly detection is a significant and hence well-studied problem. However, developing effective anomaly detection methods for complex and high-dimensional data remains a challenge. As Generative Adversarial Networks (GANs) are able to mod…
View article: Optimistic mirror descent in saddle-point problems: Going the extra (gradient) mile
Optimistic mirror descent in saddle-point problems: Going the extra (gradient) mile Open
International audience
View article: Manifold regularization with GANs for semi-supervised learning
Manifold regularization with GANs for semi-supervised learning Open
Generative Adversarial Networks are powerful generative models that are able to model the manifold of natural images. We leverage this property to perform manifold regularization by approximating a variant of the Laplacian norm using a Mon…
View article: Optimistic mirror descent in saddle-point problems: Going the extra (gradient) mile
Optimistic mirror descent in saddle-point problems: Going the extra (gradient) mile Open
Owing to their connection with generative adversarial networks (GANs), saddle-point problems have recently attracted considerable interest in machine learning and beyond. By necessity, most theoretical guarantees revolve around convex-conc…
View article: Mirror descent in saddle-point problems: Going the extra (gradient) mile.
Mirror descent in saddle-point problems: Going the extra (gradient) mile. Open
View article: Semi-Supervised Learning with GANs: Revisiting Manifold Regularization
Semi-Supervised Learning with GANs: Revisiting Manifold Regularization Open
GANS are powerful generative models that are able to model the manifold of natural images. We leverage this property to perform manifold regularization by approximating the Laplacian norm using a Monte Carlo approximation that is easily co…
View article: Efficient GAN-Based Anomaly Detection
Efficient GAN-Based Anomaly Detection Open
Generative adversarial networks (GANs) are able to model the complex highdimensional distributions of real-world data, which suggests they could be effective for anomaly detection. However, few works have explored the use of GANs for the a…