The most competitive noisy label learning methods rely on an unsupervised classification of clean and noisy samples, where samples classified as noisy are re-labelled and "MixMatched" with the clean samples.
The resulting two sets of samples are then mixed-up to create a more balanced training distribution from which a neural network can effectively learn without incurring in heavily under-fitting the minority classes.
MSACL learns representations by separating several types of strong and weak augmentations of normal image samples, where the weak augmentations represent normal images and strong augmentations denote synthetic abnormal images.
In this paper, we propose the first 3D semantic mapping system from knee arthroscopy that solves the three challenges above.
We propose probabilistic task modelling -- a generative probabilistic model for collections of tasks used in meta-learning.
Hence, we introduce a new unsupervised anomaly detection and localisation method trained exclusively with serial images that do not contain any lesion changes.
To be specific, ScanMix is designed based on the expectation maximisation (EM) framework, where the E-step estimates the value of a latent variable to cluster the training images based on their appearance representations and classification results, and the M-step optimises the SSL classification and learns effective feature representations via semantic clustering.
Ranked #11 on Image Classification on mini WebVision 1.0
Deep neural network models are robust to a limited amount of label noise, but their ability to memorise noisy labels in high noise rate problems is still an open issue.
Ranked #2 on Image Classification on Food-101N
The classification accuracy of deep learning models depends not only on the size of their training sets, but also on the quality of their labels.
Unsupervised anomaly detection (UAD) learns one-class classifiers exclusively with normal (i. e., healthy) images to detect any abnormal (i. e., unhealthy) samples that do not conform to the expected normal patterns.
In this paper, we propose Self-supervised Mean Teacher for Semi-supervised (S$^2$MTS$^2$) learning that combines self-supervised mean-teacher pre-training with semi-supervised fine-tuning.
In this paper, we introduce a new post-hoc method for OS time prediction that does not require segmentation map annotation for training.
The adversarial interpolation is enforced to consistently learn a smooth Gaussian descriptor, even when the training data is small or contaminated with anomalous samples.
Ranked #2 on Anomaly Detection on Fashion-MNIST
To address this issue, we introduce a novel and theoretically sound method, named Robust Temporal Feature Magnitude learning (RTFM), which trains a feature magnitude learning function to effectively recognise the positive instances, substantially improving the robustness of the MIL approach to the negative instances from abnormal videos.
In this paper, we propose and analyse a system that can automatically detect, localise and classify polyps from colonoscopy videos.
no code implementations • 2 Jan 2021 • Sourav Garg, Niko Sünderhauf, Feras Dayoub, Douglas Morrison, Akansel Cosgun, Gustavo Carneiro, Qi Wu, Tat-Jun Chin, Ian Reid, Stephen Gould, Peter Corke, Michael Milford
In robotics and related research fields, the study of understanding is often referred to as semantics, which dictates what does the world "mean" to a robot, and is strongly tied to the question of how to represent that meaning.
Then, we empirically show that for a large range of learning rates, SGD traverses the loss landscape across regions with largest eigenvalue of the Hessian similar to the inverse of the learning rate.
Label noise is a natural event of data collection and annotation and has been shown to have significant impact on the performance of deep learning models regarding accuracy reduction and sample complexity increase.
Domain generalisation (DG) methods address the problem of domain shift, when there is a mismatch between the distributions of training and target domains.
As deep learning models depend on correctly labeled data sets and label correctness is difficult to guarantee, it is crucial to consider the presence of noisy labels for deep learning training.
In this work, we study a new variant of the noisy label problem that combines the open-set and closed-set noisy labels, and introduce a benchmark evaluation to assess the performance of training algorithms under this setup.
Based on pseudo labels, we propose a novel unsupervised metric loss which enforces the positive concentration and negative separation of samples in the embedding space.
In this paper, we propose a novel self-supervised monocular depth estimation to regularise the training of the semantic segmentation in knee arthroscopy.
Instead, we assume that we can access a plethora of annotated image data sets from different domains (sources) and a limited number of annotated image data sets from the domain of interest (target), where each domain denotes not only different image appearance but also a different type of cell segmentation problem.
Anomaly detection methods generally target the learning of a normal image distribution (i. e., inliers showing healthy cases) and during testing, samples relatively far from the learned distribution are classified as anomalies (i. e., outliers showing disease cases).
The deployment of automated systems to diagnose diseases from medical images is challenged by the requirement to localise the diagnosed diseases to justify or explain the classification decision.
We show that the extension of the state-of-the-art self-supervised monocular trained depth estimator Monodepth2 with these two ideas allows us to design a model that produces the best results in the field in KITTI 2015 and Make3D, closing the gap with respect self-supervised stereo training and fully supervised approaches.
We introduce a new and rigorously-formulated PAC-Bayes few-shot meta-learning algorithm that implicitly learns a prior distribution of the model of interest.
We show that our proposed approach achieves the state-of-the-art result on this data set, compared with recently proposed anomaly detection systems.
However, the lack of annotations for the localisation of the regions of interest (ROIs) containing lymph nodes can limit classification accuracy due to the small size of the relevant ROIs in this problem.
Machine learning models for medical image analysis often suffer from poor performance on important subsets of a population that are not identified during training or testing.
Most state-of-the-art GZSL approaches rely on a mapping between latent visual and semantic spaces without considering if a particular sample belongs to the set of seen or unseen classes.
In the context of GZSL, semantic information is non-visual data such as a text description of both seen and unseen classes.
A solution to mitigate the small training set issue is to pre-train a denoising model with small training sets containing pairs of clean and synthesized noisy signals, produced from empirical noise priors, and fine-tune on the available small training set.
We introduce a new, rigorously-formulated Bayesian meta-learning algorithm that learns a probability distribution of model parameter prior for few-shot learning.
Meta-training has been empirically demonstrated to be the most effective pre-training method for few-shot learning of medical image classifiers (i. e., classifiers modeled with small training sets).
Deep learning models have demonstrated outstanding performance in several problems, but their training process tends to require immense amounts of computational and human resources for training and labeling, constraining the types of problems that can be tackled.
Another approach explored in the field relies on an ad-hoc linearization (in terms of N) of the triplet loss that introduces class centroids, which must be optimized using the whole training set for each mini-batch - this means that a naive implementation of this approach has run-time complexity O(N^2).
In this paper, we mitigate these issues by proposing a new GZSL method based on multi-modal training and testing processes, where the optimization explicitly promotes a balanced classification accuracy between seen and unseen classes.
We introduce Probabilistic Object Detection, the task of detecting objects in images and accurately quantifying the spatial and semantic uncertainties of the detections.
In this paper, we propose a learning method to train diagnosis models, where our approach is designed to work with relatively small datasets.
Furthermore, the proposed measurements also allow us to show that it is possible to optimise the training process with a new dynamic sampling training approach that continuously and automatically change the mini-batch size and learning rate during the training process.
Conversely, traditional approaches follow a pre-hoc approach that initially localises suspicious areas that are subsequently classified to establish the breast malignancy -- this approach is trained using strongly annotated data (i. e., it needs a delineation and classification of all lesions in an image).
In generalized zero shot learning (GZSL), the set of classes are split into seen and unseen classes, where training relies on the semantic features of the seen and unseen classes and the visual representations of only the seen classes, while testing uses the visual representations of the seen and unseen classes.
Ranked #4 on Generalized Zero-Shot Learning on SUN Attribute
There is a heated debate on how to interpret the decisions provided by deep learning models (DLM), where the main approaches rely on the visualization of salient regions to interpret the DLM classification process.
In this paper, we present a novel open-set semantic instance segmentation approach capable of segmenting all known and unknown object classes in images, based on the output of an object detector trained on known object classes.
Current approaches to explaining the decisions of deep learning systems for medical tasks have focused on visualising the elements that have contributed to each decision.
This process bears no direct resemblance with radiologist training, which is based on solving a series of tasks of increasing difficulty, where each task involves the use of significantly smaller datasets than those used in machine learning.
We developed an automated deep learning system to detect hip fractures from frontal pelvic x-rays, an important and common radiological task.
Data augmentation is an essential part of the training process applied to deep learning models.
In this paper, we propose a novel deep metric learning method that combines the triplet model and the global structure of the embedding space.
In mammography, the efficacy of computer-aided detection methods depends, in part, on the robust localisation of micro-calcifications ($\mu$C).
We propose new methods for the prediction of 5-year mortality in elderly individuals using chest computed tomography (CT).
In this work we propose a unsupervised framework to learn a deep convolutional neural network for single view depth predic- tion, without requiring a pre-training stage or annotated ground truth depths.
Current results from machine learning show that replacing this siamese by a triplet network can improve the classification accuracy in several problems, but this has yet to be demonstrated for local image descriptor learning.
We introduce two new structured output models that use a latent graph, which is flexible in terms of the number of nodes and structure, where the training process minimises a high-order loss function using a weakly annotated training set.
In this paper, we introduce a new deep convolutional neural network (ConvNet) module that promotes competition among a set of multi-scale convolutional filters.
Ranked #15 on Image Classification on MNIST
The combination of deep learning models and piecewise linear activation functions allows for the estimation of exponentially complex functions with the use of a large number of subnetworks specialized in the classification of similar input examples.
Ranked #19 on Image Classification on MNIST
Convolutional Neural Networks (ConvNets) have successfully contributed to improve the accuracy of regression-based methods for computer vision tasks such as human pose estimation, landmark localization, and object detection.
In this paper, we present a novel method for the segmentation of breast masses from mammograms exploring structured and deep learning.
In this paper, we propose a new methodology for segmenting non-rigid visual objects, where the search procedure is onducted directly on a sparse low-dimensional manifold, guided by the classification results computed from a deep belief network.
We address these two issues with an innovative structured inference using deep belief networks that produces reliable initial guess and appearance model.
In this paper, we propose the use of sparse manifolds to reduce the dimensionality of the rigid detection search space of current stateof-the-art top-down segmentation methodologies.