4.6k citations · 4.8k across the 27 of their papers we have counts for
30 papers
Ambient Physics: Training Neural PDE Solvers with Partial Observations
Harris Abdul Majid, Giannis Daras, Francesco Tudisco +1
In many scientific settings, acquiring complete observations of PDE coefficients and solutions can be expensive, hazardous, or impossible. Recent diffusion-based methods can recons…
Learning to Name Classes for Vision and Language Models
Sarah Parisot, Yongxin Yang, Steven McDonagh
Large scale vision and language models can achieve impressive zero-shot recognition performance by mapping class specific text queries to image content. Two distinct challenges tha…
Tunable Convolutions with Parametric Multi-Loss Optimization
Matteo Maggioni, Thomas Tanay, Francesca Babiloni +2
Behavior of neural networks is irremediably determined by the specific loss and data used during training. However it is often desirable to tune the model at inference time based o…
Content-Diverse Comparisons improve IQA
William Thong, Jose Costa Pereira, Sarah Parisot +2
Image quality assessment (IQA) forms a natural and often straightforward undertaking for humans, yet effective automation of the task remains highly challenging. Recent metrics fro…
Label-Efficient Object Detection via Region Proposal Network Pre-Training
Nanqing Dong, Linus Ericsson, Yongxin Yang +2
Self-supervised pre-training, based on the pretext task of instance discrimination, has fueled the recent advance in label-efficient object detection. However, existing studies foc…
CLAD: A realistic Continual Learning benchmark for Autonomous Driving
Eli Verwimp, Kuo Yang, Sarah Parisot +5
In this paper we describe the design and the ideas motivating a new Continual Learning benchmark for Autonomous Driving (CLAD), that focuses on the problems of object classificatio…
Out-of-Distribution Detection with Class Ratio Estimation
Mingtian Zhang, Andi Zhang, Tim Z. Xiao +2
Density-based Out-of-distribution (OOD) detection has recently been shown unreliable for the task of detecting OOD images. Various density ratio based approaches achieve good empir…
Re-examining Distillation For Continual Object Detection
Eli Verwimp, Kuo Yang, Sarah Parisot +5
Training models continually to detect and classify objects, from new classes and new domains, remains an open problem. In this work, we conduct a thorough analysis of why and how o…
CroMo: Cross-Modal Learning for Monocular Depth Estimation
Yannick Verdié, Jifei Song, Barnabé Mas +3
Learning-based depth estimation has witnessed recent progress in multiple directions; from self-supervision using monocular video to supervised methods offering highest accuracy. C…
Model-Based Image Signal Processors via Learnable Dictionaries
Marcos V. Conde, Steven McDonagh, Matteo Maggioni +2
Digital cameras transform sensor RAW readings into RGB images by means of their Image Signal Processor (ISP). Computational photography tasks such as image denoising and colour con…
Long-tail Recognition via Compositional Knowledge Transfer
Sarah Parisot, Pedro M. Esperanca, Steven McDonagh +3
In this work, we introduce a novel strategy for long-tail recognition that addresses the tail classes' few-shot problem via training-free knowledge transfer. Our objective is to tr…
Spread Flows for Manifold Modelling
Mingtian Zhang, Yitong Sun, Chen Zhang +1
Flow-based models typically define a latent space with dimensionality identical to the observational space. In many problems, however, the data does not populate the full ambient d…
On the Out-of-distribution Generalization of Probabilistic Image Modelling
Mingtian Zhang, Andi Zhang, Steven McDonagh
Out-of-distribution (OOD) detection and lossless compression constitute two problems that can be solved by the training of probabilistic models on a first dataset with subsequent l…
NTIRE 2021 Challenge on Perceptual Image Quality Assessment
Jinjin Gu, Haoming Cai, Chao Dong +47
This paper reports on the NTIRE 2021 challenge on perceptual image quality assessment (IQA), held in conjunction with the New Trends in Image Restoration and Enhancement workshop (…
Residual Contrastive Learning for Image Reconstruction: Learning Transferable Representations from Noisy Images
Nanqing Dong, Matteo Maggioni, Yongxin Yang +3
This paper is concerned with contrastive learning (CL) for low-level image restoration and enhancement tasks. We propose a new label-efficient learning paradigm based on residuals,…
Many-shot from Low-shot: Learning to Annotate using Mixed Supervision for Object Detection
Carlo Biffi, Steven McDonagh, Philip Torr +2
Object detection has witnessed significant progress by relying on large, manually annotated datasets. Annotating such datasets is highly time consuming and expensive, which motivat…
Low Light Video Enhancement using Synthetic Data Produced with an Intermediate Domain Mapping
Danai Triantafyllidou, Sean Moran, Steven McDonagh +2
Advances in low-light video RAW-to-RGB translation are opening up the possibility of fast low-light imaging on commodity devices (e.g. smartphone cameras) without the need for a tr…
DeepLPF: Deep Local Parametric Filters for Image Enhancement
Sean Moran, Pierre Marza, Steven McDonagh +2
Digital artists often improve the aesthetic quality of digital photographs through manual retouching. Beyond global adjustments, professional image editing programs provide local a…
A Multi-Hypothesis Approach to Color Constancy
Daniel Hernandez-Juarez, Sarah Parisot, Benjamin Busam +3
Contemporary approaches frame the color constancy problem as learning camera specific illuminant mappings. While high accuracy can be achieved on camera specific data, these models…
SteReFo: Efficient Image Refocusing with Stereo Vision
Benjamin Busam, Matthieu Hog, Steven McDonagh +1
Whether to attract viewer attention to a particular object, give the impression of depth or simply reproduce human-like scene perception, shallow depth of field images are used ext…
Formulating Camera-Adaptive Color Constancy as a Few-shot Meta-Learning Problem
Steven McDonagh, Sarah Parisot, Fengwei Zhou +4
Digital camera pipelines employ color constancy methods to estimate an unknown scene illuminant, in order to re-illuminate images as if they were acquired under an achromatic light…
EchoFusion: Tracking and Reconstruction of Objects in 4D Freehand Ultrasound Imaging without External Trackers
Bishesh Khanal, Alberto Gomez, Nicolas Toussaint +11
Ultrasound (US) is the most widely used fetal imaging technique. However, US images have limited capture range, and suffer from view dependent artefacts such as acoustic shadows. C…
Automatic View Planning with Multi-scale Deep Reinforcement Learning Agents
Amir Alansary, Loic Le Folgoc, Ghislain Vaillant +11
We propose a fully automatic method to find standardized view planes in 3D image acquisitions. Standard view images are important in clinical practice as they provide a means to pe…
Attention U-Net: Learning Where to Look for the Pancreas
Ozan Oktay, Jo Schlemper, Loic Le Folgoc +9
We propose a novel attention gate (AG) model for medical imaging that automatically learns to focus on target structures of varying shapes and sizes. Models trained with AGs implic…