1.5k citations · 2.7k across the 37 of their papers we have counts for
45 papers
Defensive Tensorization
Adrian Bulat, Jean Kossaifi, Sourav Bhattacharya +5
We propose defensive tensorization, an adversarial defence technique that leverages a latent high-order factorization of the network. The layers of a network are first expressed as…
SAIC_Cambridge-HuPBA-FBK Submission to the EPIC-Kitchens-100 Action Recognition Challenge 2021
Swathikiran Sudhakaran, Adrian Bulat, Juan-Manuel Perez-Rua +5
This report presents the technical details of our submission to the EPIC-Kitchens-100 Action Recognition Challenge 2021. To participate in the challenge we deployed spatio-temporal…
WarpedGANSpace: Finding non-linear RBF paths in GAN latent space
Christos Tzelepis, Georgios Tzimiropoulos, Ioannis Patras
This work addresses the problem of discovering, in an unsupervised manner, interpretable paths in the latent space of pretrained GANs, so as to provide an intuitive and easy way of…
Space-time Mixing Attention for Video Transformer
Adrian Bulat, Juan-Manuel Perez-Rua, Swathikiran Sudhakaran +2
This paper is on video recognition using Transformers. Very recent attempts in this area have demonstrated promising results in terms of recognition accuracy, yet they have been al…
Bit-Mixer: Mixed-precision networks with runtime bit-width selection
Adrian Bulat, Georgios Tzimiropoulos
Mixed-precision networks allow for a variable bit-width quantization for every layer in the network. A major limitation of existing work is that the bit-width for each layer must b…
Affective Processes: stochastic modelling of temporal context for emotion and facial expression recognition
Enrique Sanchez, Mani Kumar Tellamekala, Michel Valstar +1
Temporal context is key to the recognition of expressions of emotion. Existing methods, that rely on recurrent or self-attention models to enforce temporal consistency, work on the…
Pre-training strategies and datasets for facial representation learning
Adrian Bulat, Shiyang Cheng, Jing Yang +3
What is the best way to learn a universal face representation? Recent work on Deep Learning in the area of face analysis has focused on supervised learning for specific tasks of in…
Improving memory banks for unsupervised learning with large mini-batch, consistency and hard negative mining
Adrian Bulat, Enrique Sánchez-Lozano, Georgios Tzimiropoulos
An important component of unsupervised learning by instance-based discrimination is a memory bank for storing a feature representation for each training sample in the dataset. In t…
Semi-supervised Facial Action Unit Intensity Estimation with Contrastive Learning
Enrique Sanchez, Adrian Bulat, Anestis Zaganidis +1
This paper tackles the challenging problem of estimating the intensity of Facial Action Units with few labeled images. Contrary to previous works, our method does not require to ma…
High-Capacity Expert Binary Networks
Adrian Bulat, Brais Martinez, Georgios Tzimiropoulos
Network binarization is a promising hardware-aware direction for creating efficient deep models. Despite its memory and computational advantages, reducing the accuracy gap between…
A Transfer Learning approach to Heatmap Regression for Action Unit intensity estimation
Ioanna Ntinou, Enrique Sanchez, Adrian Bulat +2
Action Units (AUs) are geometrically-based atomic facial muscle movements known to produce appearance changes at specific facial locations. Motivated by this observation we propose…
Training Binary Neural Networks with Real-to-Binary Convolutions
Brais Martinez, Jing Yang, Adrian Bulat +1
This paper shows how to train binary networks to within a few percent points () of the full precision counterpart. We first show how to build a strong baseline, which…
Knowledge distillation via adaptive instance normalization
Jing Yang, Brais Martinez, Adrian Bulat +1
This paper addresses the problem of model compression via knowledge distillation. To this end, we propose a new knowledge distillation method based on transferring feature statisti…
BATS: Binary ArchitecTure Search
Adrian Bulat, Brais Martinez, Georgios Tzimiropoulos
This paper proposes Binary ArchitecTure Search (BATS), a framework that drastically reduces the accuracy gap between binary neural networks and their real-valued counterparts by me…
Toward fast and accurate human pose estimation via soft-gated skip connections
Adrian Bulat, Jean Kossaifi, Georgios Tzimiropoulos +1
This paper is on highly accurate and highly efficient human pose estimation. Recent works based on Fully Convolutional Networks (FCNs) have demonstrated excellent results for this…
Towards Pose-invariant Lip-Reading
Shiyang Cheng, Pingchuan Ma, Georgios Tzimiropoulos +4
Lip-reading models have been significantly improved recently thanks to powerful deep learning architectures. However, most works focused on frontal or near frontal views of the mou…
Object landmark discovery through unsupervised adaptation
Enrique Sanchez, Georgios Tzimiropoulos
This paper proposes a method to ease the unsupervised learning of object landmark detectors. Similarly to previous methods, our approach is fully unsupervised in a sense that it do…
XNOR-Net++: Improved Binary Neural Networks
Adrian Bulat, Georgios Tzimiropoulos
This paper proposes an improved training algorithm for binary neural networks in which both weights and activations are binary numbers. A key but fairly overlooked feature of the c…
AnimalWeb: A Large-Scale Hierarchical Dataset of Annotated Animal Faces
Muhammad Haris Khan, John McDonagh, Salman Khan +5
Being heavily reliant on animals, it is our ethical obligation to improve their well-being by understanding their needs. Several studies show that animal needs are often expressed…
Matrix and tensor decompositions for training binary neural networks
Adrian Bulat, Jean Kossaifi, Georgios Tzimiropoulos +1
This paper is on improving the training of binary neural networks in which both activations and weights are binary. While prior methods for neural network binarization binarize eac…
Improved training of binary networks for human pose estimation and image recognition
Adrian Bulat, Georgios Tzimiropoulos, Jean Kossaifi +1
Big neural networks trained on large datasets have advanced the state-of-the-art for a large variety of challenging problems, improving performance by a large margin. However, unde…
T-Net: Parametrizing Fully Convolutional Nets with a Single High-Order Tensor
Jean Kossaifi, Adrian Bulat, Georgios Tzimiropoulos +1
Recent findings indicate that over-parametrization, while crucial for successfully training deep neural networks, also introduces large amounts of redundancy. Tensor methods have t…
Incremental multi-domain learning with network latent tensor factorization
Adrian Bulat, Jean Kossaifi, Georgios Tzimiropoulos +1
The prominence of deep learning, large amount of annotated data and increasingly powerful hardware made it possible to reach remarkable performance for supervised classification ta…
Features Extraction Based on an Origami Representation of 3D Landmarks
Juan Manuel Fernandez Montenegro, Mahdi Maktab Dar Oghaz, Athanasios Gkelias +2
Feature extraction analysis has been widely investigated during the last decades in computer vision community due to the large range of possible applications. Significant work has…