Search Results for author: Gemma Roig

Found 25 papers, 11 papers with code

Predicting emotion from music videos: exploring the relative contribution of visual and auditory information to affective responses

1 code implementation19 Feb 2022 Phoebe Chua, Dimos Makris, Dorien Herremans, Gemma Roig, Kat Agres

In this paper we present MusicVideos (MuVi), a novel dataset for affective multimedia content analysis to study how the auditory and visual modalities contribute to the perceived emotion of media.

Feature Importance Multimodal Emotion Recognition +1

FRIDA -- Generative Feature Replay for Incremental Domain Adaptation

no code implementations28 Dec 2021 Sayan Rakshit, Anwesh Mohanty, Ruchika Chavhan, Biplab Banerjee, Gemma Roig, Subhasis Chaudhuri

Inspired by the notion of generative feature replay, we propose a novel framework called Feature Replay based Incremental Domain Adaptation (FRIDA) which leverages a new incremental generative adversarial network (GAN) called domain-generic auxiliary classification GAN (DGAC-GAN) for producing domain-specific feature representations seamlessly.

Unsupervised Domain Adaptation

AttendAffectNet–Emotion Prediction of Movie Viewers Using Multimodal Fusion with Self-Attention

1 code implementation Sensors 2021 Ha Thi Phuong Thao, B T Balamurali, Gemma Roig, Dorien Herremans

The models that use all visual, audio, and text features simultaneously as their inputs performed better than those using features extracted from each modality separately.

Representation Learning

AttendAffectNet: Self-Attention based Networks for Predicting Affective Responses from Movies

1 code implementation21 Oct 2020 Ha Thi Phuong Thao, Balamurali B. T., Dorien Herremans, Gemma Roig

In this work, we propose different variants of the self-attention based network for emotion prediction from movies, which we call AttendAffectNet.

Duality Diagram Similarity: a generic framework for initialization selection in task transfer learning

2 code implementations ECCV 2020 Kshitij Dwivedi, Jiahui Huang, Radoslaw Martin Cichy, Gemma Roig

In this paper, we tackle an open research question in transfer learning, which is selecting a model initialization to achieve high performance on a new task, given several pre-trained models.

Model Selection Semantic Segmentation +1

Using Human Psychophysics to Evaluate Generalization in Scene Text Recognition Models

no code implementations30 Jun 2020 Sahar Siddiqui, Elena Sizikova, Gemma Roig, Najib J. Majaj, Denis G. Pelli

Relative to the attention-based (Attn) model, we discover that the connectionist temporal classification (CTC) model is more robust to noise and occlusion, and better at generalizing to different word lengths.

Scene Text Recognition

LCD: Learned Cross-Domain Descriptors for 2D-3D Matching

1 code implementation21 Nov 2019 Quang-Hieu Pham, Mikaela Angelina Uy, Binh-Son Hua, Duc Thanh Nguyen, Gemma Roig, Sai-Kit Yeung

In this work, we present a novel method to learn a local cross-domain descriptor for 2D image and 3D point cloud matching.

3D Point Cloud Matching Depth Estimation

Predictive Coding Networks Meet Action Recognition

no code implementations22 Oct 2019 Xia Huang, Hossein Mousavi, Gemma Roig

In this way, the model only relies on the video frames, and does not need pre-processed optical flows as input.

Action Recognition Optical Flow Estimation

Latent space representation for multi-target speaker detection and identification with a sparse dataset using Triplet neural networks

1 code implementation1 Oct 2019 Kin Wai Cheuk, Balamurali B. T., Gemma Roig, Dorien Herremans

When reducing the training data to only using the train set, our method results in 309 confusions for the Multi-target speaker identification task, which is 46% better than the baseline model.

Speaker Identification Speaker Recognition

Multimodal Deep Models for Predicting Affective Responses Evoked by Movies

1 code implementation16 Sep 2019 Ha Thi Phuong Thao, Dorien Herremans, Gemma Roig

Interestingly, we also observe that the optical flow is more informative than the RGB in videos, and overall, models using audio features are more accurate than those based on video features when making the final prediction of evoked emotions.

Optical Flow Estimation

Representation Similarity Analysis for Efficient Task taxonomy & Transfer Learning

2 code implementations CVPR 2019 Kshitij Dwivedi, Gemma Roig

We next evaluate the relationship of RSA with the transfer learning performance on Taskonomy tasks and a new task: Pascal VOC semantic segmentation.

Semantic Segmentation Transfer Learning

Deep Anchored Convolutional Neural Networks

no code implementations22 Apr 2019 Jiahui Huang, Kshitij Dwivedi, Gemma Roig

Convolutional Neural Networks (CNNs) have been proven to be extremely successful at solving computer vision tasks.

DOPING: Generative Data Augmentation for Unsupervised Anomaly Detection with GAN

no code implementations23 Aug 2018 Swee Kiat Lim, Yi Loo, Ngoc-Trung Tran, Ngai-Man Cheung, Gemma Roig, Yuval Elovici

To the best of our knowledge, our method is the first data augmentation technique focused on improving performance in unsupervised anomaly detection.

Data Augmentation Unsupervised Anomaly Detection

Do Deep Neural Networks Suffer from Crowding?

2 code implementations NeurIPS 2017 Anna Volokitin, Gemma Roig, Tomaso Poggio

Also, for all tested networks, when trained on targets in isolation, we find that recognition accuracy of the networks decreases the closer the flankers are to the target and the more flankers there are.

Object Recognition

Herding Generalizes Diverse M -Best Solutions

no code implementations14 Nov 2016 Ece Ozkan, Gemma Roig, Orcun Goksel, Xavier Boix

We show that the algorithm to extract diverse M -solutions from a Conditional Random Field (called divMbest [1]) takes exactly the form of a Herding procedure [2], i. e. a deterministic dynamical system that produces a sequence of hypotheses that respect a set of observed moment constraints.

Semantic Segmentation

Foveation-based Mechanisms Alleviate Adversarial Examples

no code implementations19 Nov 2015 Yan Luo, Xavier Boix, Gemma Roig, Tomaso Poggio, Qi Zhao

To see this, first, we report results in ImageNet that lead to a revision of the hypothesis that adversarial perturbations are a consequence of CNNs acting as a linear classifier: CNNs act locally linearly to changes in the image regions with objects recognized by the CNN, and in other regions the CNN may act non-linearly.

Foveation Translation

Self-Adaptable Templates for Feature Coding

no code implementations NeurIPS 2014 Xavier Boix, Gemma Roig, Salomon Diether, Luc V. Gool

Within this processing pipeline, the common trend is to learn the feature coding templates, often referred as codebook entries, filters, or over-complete basis.

Image Classification Object Recognition +1

Comment on "Ensemble Projection for Semi-supervised Image Classification"

no code implementations29 Aug 2014 Xavier Boix, Gemma Roig, Luc van Gool

In a series of papers by Dai and colleagues [1, 2], a feature map (or kernel) was introduced for semi- and unsupervised learning.

Classification General Classification +1

SEEDS: Superpixels Extracted via Energy-Driven Sampling

1 code implementation16 Sep 2013 Michael Van den Bergh, Xavier Boix, Gemma Roig, Luc van Gool

We define a robust and fast to evaluate energy function, based on enforcing color similarity between the bound- aries and the superpixel color histogram.


Random Binary Mappings for Kernel Learning and Efficient SVM

no code implementations19 Jul 2013 Gemma Roig, Xavier Boix, Luc van Gool

SVMs suffer from various drawbacks in terms of selecting the right kernel, which depends on the image descriptors, as well as computational and memory efficiency.


Cannot find the paper you are looking for? You can Submit a new open access paper.