Publications related to Semi-supervised Active Salient Object Detection

A graph convolutional autoencoder approach to model order reduction for parametrized PDEs

The present work proposes a framework for nonlinear model order reduction based on a Graph Convolutional Autoencoder (GCA-ROM). In the reduced order modeling (ROM) context, one is interested in obtaining real -time and many-query evaluations of parametric ...

Academic Press Inc Elsevier Science2024

Neural Distributed Image Compression with Cross-Attention Feature Alignment

Ali Garjani

We consider the problem of compressing an information source when a correlated one is available as side information only at the decoder side, which is a special case of the distributed source coding problem in information theory. In particular, we consider ...

IEEE COMPUTER SOC2023

Automatic Parameterization for Aerodynamic Shape Optimization via Deep Geometric Learning

Pascal Fua, Zhen Wei

We propose two deep learning models that fully automate shape parameterization for aerodynamic shape optimization. Both models are optimized to parameterize via deep geometric learning to embed human prior knowledge into learned geometric patterns, elimina ...

2023

Sparse Autoencoders for Speech Modeling and Recognition

Selen Hande Kabil

Speech recognition-based applications upon the advancements in artificial intelligence play an essential role to transform most aspects of modern life. However, speech recognition in real-life conditions (e.g., in the presence of overlapping speech, varyin ...

EPFL2023

Towards learning-based denoising of light fields

Touradj Ebrahimi, Michela Testolina, Tomás Soares De Carvalho Feith

In recent years, new emerging immersive imaging modalities, e.g. light fields, have been receiving growing attention, becoming increasingly widespread over the years. Light fields are often captured through multi-camera arrays or plenoptic cameras, with th ...

2023

RG-Flow: a hierarchical and explainable flow model based on renormalization group and sparse prior

Dian Wu

Flow-based generative models have become an important class of unsupervised learning approaches. In this work, we incorporate the key ideas of renormalization group (RG) and sparse prior distribution to design a hierarchical flow-based generative model, RG ...

IOP Publishing Ltd2022

The spectral bias of polynomial neural networks

Volkan Cevher, Moulik Choraria, Leello Tadesse Dadi, Grigorios Chrysos

Polynomial neural networks (PNNs) have been recently shown to be particularly effective at image generation and face recognition, where high-frequency information is critical. Previous studies have revealed that neural networks demonstrate a spectral bias ...

2022

Self-Supervision By Prediction For Object Discovery In Videos

Pascal Frossard, Beril Besbinar

Despite their irresistible success, deep learning algorithms still heavily rely on annotated data, and unsupervised settings pose many challenges, such as finding the right inductive bias in diverse scenarios. In this paper, we propose an object-centric mo ...

IEEE2021

Fidelity Estimation Improves Noisy-Image Classification with Pretrained Networks

Sabine Süsstrunk, Majed El Helou, Deblina Bhattacharjee, Xiaoyu Lin

Image classification has significantly improved using deep learning. This is mainly due to convolutional neural networks (CNNs) that are capable of learning rich feature extractors from large datasets. However, most deep learning classification methods are ...

IEEE Institute of Electrical and Electronics Engineers2021

Learning-based Image Coding: Early Solutions Reviewing and Subjective Quality Evaluation

Touradj Ebrahimi, Pinar Akyazi

Nowadays, image and video are the data types that consume most of the resources of modern communication channels, both in fixed and wireless networks. Thus, it is vital to compress visual data as much as possible, while maintaining some target quality leve ...

SPIE-INT SOC OPTICAL ENGINEERING2021

Deep Image Restoration: Between Data Fidelity and Learned Priors

Majed El Helou

Image restoration reconstructs, as faithfully as possible, an original image from a potentially degraded version of it. Image degradations can be of various types, for instance haze, unwanted reflections, optical or spectral aberrations, or other physicall ...

EPFL2021

STC-GAN: Spatio-Temporally Coupled Generative Adversarial Networks for Predictive Scene Parsing

Jie Luo, Yiyu Wang, Mengshi Qi

Predictive scene parsing is a task of assigning pixel-level semantic labels to a future frame of a video. It has many applications in vision-based artificial intelligent systems, e.g., autonomous driving and robot navigation. Although previous work has sho ...

IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC2020

AL2: Progressive Activation Loss for Learning General Representations in Classification Neural Networks

Sabine Süsstrunk, Frederike Dümbgen, Majed El Helou

The large capacity of neural networks enables them to learn complex functions. To avoid overfitting, networks however require a lot of training data that can be expensive and time-consuming to collect. A common practical approach to attenuate overfitting i ...

IEEE2020

Concept Discovery for The Interpretation of Landscape Scenicness

Devis Tuia

In this paper, we study how to extract visual concepts to understand landscape scenicness. Using visual feature representations from a Convolutional Neural Network (CNN), we learn a number of Concept Activation Vectors (CAV) aligned with semantic concepts ...

2020

Learning stereo reconstruction with deep neural networks

Stepan Tulyakov

Stereo reconstruction is a problem of recovering a 3d structure of a scene from a pair of images of the scene, acquired from different viewpoints. It has been investigated for decades and many successful methods were developed. The main drawback of these ...

EPFL2020

Evaluating and Interpreting Deep Convolutional Neural Networks via Non-negative Matrix Factorization

Edo Collins

With ever greater computational resources and more accessible software, deep neural networks have become ubiquitous across industry and academia. Their remarkable ability to generalize to new samples defies the conventional view, which holds that complex, ...

EPFL2019

Full-Gradient Representation for Neural Network Visualization

François Fleuret, Suraj Srinivas

We introduce a new tool for interpreting neural net responses, namely full-gradients, which decomposes the neural net response into input sensitivity and per-neuron sensitivity components. This is the first proposed representation which satisfies two key p ...

2019

Learning Robust Features and Latent Representations for Single View 3D Pose Estimation of Humans and Objects

Bugra Tekin

Estimating the 3D poses of rigid and articulated bodies is one of the fundamental problems of Computer Vision. It has a broad range of applications including augmented reality, surveillance, animation and human-computer interaction. Despite the ever-growin ...

EPFL2018

Investigating Depth Domain Adaptation for Efficient Human Pose Estimation

Jean-Marc Odobez, Olivier Canévet, Michael Villamizar

Convolutional Neural Networks (CNN) are the leading models for human body landmark detection from RGB vision data. However, as such models require high computational load, an alternative is to rely on depth images which, due to their more simple nature, ca ...

SPRINGER INTERNATIONAL PUBLISHING AG2018

Denoising and Raw-waveform Networks for Weakly-Supervised Gender Identification on Noisy Speech

Mathew Magimai Doss, Manoj Kumar, Jilt Sebastian

This paper presents a raw-waveform neural network and uses it along with a denoising network for clustering in weakly supervised learning scenarios under extreme noise conditions. Specifically, we consider language independent Automatic Gender Recognition ...

ISCA-INT SPEECH COMMUNICATION ASSOC2018