Learning to Represent and Reconstruct 3D Deformable Objects

Representing and reconstructing 3D deformable shapes are two tightly linked problems that have long been studied within the computer vision field. Deformable shapes are truly ubiquitous in the real world, whether be it specific object classes such as humans, garments and animals or more abstract ones such as generic materials deforming under an external force. Practical computer vision algorithms must be able to understand the shapes of objects in the observed scenes to unlock the wide spectrum of much sought after applications ranging from virtual try-on to automated surgeries.Automatic shape reconstruction is known to be an ill-posed problem, especially in the common scenario of a single image input. Therefore, the modern approaches rely on deep learning paradigm which has proven to be extremely effective even for the severely under-constrained computer vision problems. We, too, exploit the success of data-driven approaches, however, we also show that generic deep learning models can greatly benefit from being combined with explicit knowledge originating in computational geometry. We analyze the use of various 3D shape representations and we distinctly focus on one of them, the atlas-based representation, which turns out to be especially suitable for modeling deformable shapes and which we further improve and extend.The atlas-based representation models the surfaces as an ensemble of continuous functions and thus allows for arbitrary resolution and analytical surface analysis. We identify its major shortcomings, namely the patch collapse, patch overlap and strong mapping distortions, and we propose novel regularizers based on analytically computed properties of the reconstructed surfaces. Our approach counteracts the aforementioned drawbacks while yielding higher reconstruction accuracy.We dive into the problematics of atlas-based shape representation deeper and focus on another design flaw, the global inconsistency of the mappings. While it is not reflected in quantitative metrics, it is detrimental to the visual quality of the reconstructed surfaces. Specifically, we design loss functions encouraging intercommunication among the mappings which pushes the resulting surface towards a C1 smooth function and thus dramatically improves the visual quality.Furthermore, we adapt the atlas-based representation so that it could model a full sequence of a deforming object in a temporally-consistent way. The goal is to produce such reconstruction where each surface point always represents the same semantic point on the target GT surface. To achieve such behavior, we note that if each surface point deforms close-to-isometrically, its semantic location likely remains unchanged. Practically, we make use of the Riemannian metric, and force it to remain point-wise constant throughout the sequence. The experiments show that our method yields SotA results on correspondence estimation task.Finally, we look into a particular problem of monocular texture-less deformable shape reconstruction. We propose a multi-task learning approach which jointly produces a normal map, a depth map and a mesh corresponding to the observed surface. We show that producing multiple different 3D representations of the same objects results in higher reconstruction quality. We acquire a large real-world annotated dataset of texture-less deforming objects and we release it for public use.

Mean-Field methods for Structured Deep-Learning in Computer Vision

Pierre Bruno Baqué

In recent years, Machine Learning based Computer Vision techniques made impressive progress. These algorithms proved particularly efficient for image classification or detection of isolated objects. From a probabilistic perspective, these methods can predict marginals, over single or multiple variables, independently, with high accuracy. However, in many tasks of practical interest, we need to predict jointly several correlated variables. Practical applications include people detection in crowded scenes, image segmentation, surface reconstruction, 3D pose estimation and others. A large part of the research effort in today's computer-vision community aims at finding task-specific solutions to these problems, while leveraging the power of Deep-Learning based classifiers. In this thesis, we present our journey towards a generic and practical solution based on mean-field (MF) inference. Mean-field is a Statistical Physics-inspired method which has long been used in Computer-Vision as a variational approximation to posterior distributions over complex Conditional Random Fields. Standard mean-field optimization is based on coordinate descent and in many situations can be impractical. We therefore propose a novel proximal gradient-based approach to optimizing the variational objective. It is naturally parallelizable and easy to implement. We prove its convergence, and then demonstrate that, in practice, it yields faster convergence and often finds better optima than more traditional mean-field optimization techniques. Then, we show that we can replace the fully factorized distribution of mean-field by a weighted mixture of such distributions, that similarly minimizes the KL-Divergence to the true posterior. Our extension of the clamping method proposed in previous works allows us to both produce a more descriptive approximation of the true posterior and, inspired by the diverse MAP paradigms, fit a mixture of mean-field approximations. We demonstrate that this positively impacts real-world algorithms that initially relied on mean-fields. One of the important properties of the mean-field inference algorithms is that the closed-form updates are fully differentiable operations. This naturally allows to do parameter learning by simply unrolling multiple iterations of the updates, the so-called back-mean-field algorithm. We derive a novel and efficient structured learning method for multi-modal posterior distribution based on the Multi-Modal Mean-Field approximation, which can be seamlessly combined to modern gradient-based learning methods such as CNNs. Finally, we explore in more details the specific problem of structured learning and prediction for multiple-people detection in crowded scenes. We then present a mean-field based structured deep-learning detection algorithm that provides state of the art results on this dataset.

EPFL2018

Graph-based image representation learning

Renata Khasanova

Though deep learning (DL) algorithms are very powerful for image processing tasks, they generally require a lot of data to reach their full potential. Furthermore, there is no straightforward way to impose various properties, given by the prior knowledge about the target task, on the features extracted by a DL model. Therefore, in this thesis we propose several techniques that rely on the power of graph representations to embed prior knowledge inside the learning process. This allows to reduce the solution space and leads to faster optimization convergence and higher accuracy in the representation learning. In our first work, inspired by the ability of a human to correctly classify rotated, shifted or flipped objects, we propose an algorithm that permits to inherently encode invariance to isometric transformations of objects in an image. Our DL architecture is based on graph representations and consists of three novel layers, which we refer to as graph convolutional, dynamic pooling and statistical layers. Our experiments on the image classification tasks show that our network correctly recognizes isometrically transformed objects even though such types of transformation are not seen by the network at training time. Standard DL techniques are typically not able to succeed in solving such a problem without extensive data augmentation. Then, we propose to exploit the properties of graph-based approaches to efficiently process images with various types of projective geometry. In particular, we are interested in increasingly popular omnidirectional cameras, which have a 360 degree field of view. Despite their effectiveness, such cameras create images with specific geometric properties, which require special techniques for efficient processing. We propose an efficient way of adjusting the weights of the graph edges to adapt the filter responses to the geometric image properties introduced by omnidirectional cameras. Our experiments prove that using the proposed graph with properly adjusted edge weights permits to reach better performance as compared to using regular grid graph with equal weights. Finally, the approach described above relies on the isotropic filters, which work well within our transformation invariant architecture for image classification. However, for other problems (e.g. image compression) or even when used without dynamic pooling and statistical layers that are defined within the proposed architecture, these filters are unable to efficiently encode the information about the object. Thus, we introduce a different technique based on anisotropic filters that adapt their shape and size according to the omnidirectional image geometry. The main advantage of this approach compared to the previous one is the ability to encode the orientation of an image pattern, which is important for various tasks such as image compression. Our experiments show that our approach adapts to different image projective geometries and achieves state-of-the-art performance on image classification and compression tasks. Overall we propose several methods, which combine the power of DL and graph signal processing towards incorporating prior information about the target task inside the optimization procedure. We hope that the research efforts presented in this thesis will help the development of efficient DL algorithms that can use various types of prior knowledge to make them efficient even when the available training data is scarce.

EPFL2019

Machine learning for information retrieval

David Grangier

In this thesis, we explore the use of machine learning techniques for information retrieval. More specifically, we focus on ad-hoc retrieval, which is concerned with searching large corpora to identify the documents relevant to user queries. This identification is performed through a ranking task. Given a user query, an ad-hoc retrieval system ranks the corpus documents, so that the documents relevant to the query ideally appear above the others. In a machine learning framework, we are interested in proposing learning algorithms that can benefit from limited training data in order to identify a ranker likely to achieve high retrieval performance over unseen documents and queries. This problem presents novel challenges compared to traditional learning tasks, such as regression or classification. First, our task is a ranking problem, which means that the loss for a given query cannot be measured as a sum of an individual loss suffered for each corpus document. Second, most retrieval queries present a highly unbalanced setup, with a set of relevant documents accounting only for a very small fraction of the corpus. Third, ad-hoc retrieval corresponds to a kind of "double" generalization problem, since the learned model should not only generalize to new documents but also to new queries. Finally, our task also presents challenging efficiency constraints, since ad-hoc retrieval is typically applied to large corpora. The main objective of this thesis is to investigate the discriminative learning of ad-hoc retrieval models. For that purpose, we propose different models based on kernel machines or neural networks adapted to different retrieval contexts. The proposed approaches rely on different online learning algorithms that allow efficient learning over large corpora. The first part of the thesis focuses on text retrieval. In this case, we adopt a classical approach to the retrieval ranking problem, and order the text documents according to their estimated similarity to the text query. The assessment of semantic similarity between text items plays a key role in that setup and we propose a learning approach to identify an effective measure of text similarity. This identification is not performed relying on a set of queries with their corresponding relevant document sets, since such data are especially expensive to label and hence rare. Instead, we propose to rely on hyperlink data, since hyperlinks convey semantic proximity information that is relevant to similarity learning. This setup is hence a transfer learning setup, where we benefit from the proximity information encoded by hyperlinks to improve the performance over the ad-hoc retrieval task. We then investigate another retrieval problem, i.e. the retrieval of images from text queries. Our approach introduces a learning procedure optimizing a criterion related to the ranking performance. This criterion adapts our previous learning objective for learning textual similarity to the image retrieval problem. This yields an image ranking model that addresses the retrieval problem directly. This approach contrasts with previous research that relies on an intermediate image annotation task. Moreover, our learning procedure builds upon recent work on the online learning of kernel-based classifiers. This yields an efficient, scalable algorithm, which can benefit from recent kernels developed for image comparison. In the last part of the thesis, we show that the objective function used in the previous retrieval problems can be applied to the task of keyword spotting, i.e. the detection of given keywords in speech utterances. For that purpose, we formalize this problem as a ranking task: given a keyword, the keyword spotter should order the utterances so that the utterances containing the keyword appear above the others. Interestingly, this formulation yields an objective directly maximizing the area under the receiver operating curve, the most common keyword spotter evaluation measure. This objective is then used to train a model adapted to this intrinsically sequential problem. This model is then learned with a procedure derived from the algorithm previously introduced for the image retrieval task. To conclude, this thesis introduces machine learning approaches for ad-hoc retrieval. We propose learning models for various multi-modal retrieval setups, i.e. the retrieval of text documents from text queries, the retrieval of images from text queries and the retrieval of speech recordings from written keywords. Our approaches rely on discriminative learning and enjoy efficient training procedures, which yields effective and scalable models. In all cases, links with prior approaches were investigated and experimental comparisons were conducted.

EPFL2008