Share This Author
SalGAN: Visual Saliency Prediction with Generative Adversarial Networks
This work introduces SalGAN, a deep convolutional neural network for visual saliency prediction trained with adversarial examples and shows how adversarial training allows reaching state-of-the-art performance across different metrics when combined with a widely-used loss function like BCE.
Shallow and Deep Convolutional Networks for Saliency Prediction
- Junting Pan, E. Sayrol, Xavier Giro-i-Nieto, Kevin McGuinness, N. O’Connor
- Computer ScienceIEEE Conference on Computer Vision and Pattern…
- 2 March 2016
This paper addresses the problem with a completely data-driven approach by training a convolutional neural network (convnet) and proposes two designs: a shallow convnet trained from scratch, and a another deeper solution whose first three layers are adapted from another network trained for classification.
Unsupervised label noise modeling and loss correction
- Eric Arazo Sanchez, Diego Ortego, Paul Albert, N. O’Connor, Kevin McGuinness
- Computer ScienceICML
- 25 April 2019
A suitable two-component mixture model is suggested as an unsupervised generative model of sample loss values during training to allow online estimation of the probability that a sample is mislabelled and correct the loss by relying on the network prediction.
A comparative evaluation of interactive segmentation algorithms
A multiscale representation method for nonrigid shapes with a single closed contour
The criteria that should be satisfied by a descriptor for nonrigid shapes with a single closed contour are discussed and a shape representation method that fulfills these criteria is proposed that is very efficient and invariant to several kinds of transformations.
Pseudo-Labeling and Confirmation Bias in Deep Semi-Supervised Learning
- Eric Arazo, Diego Ortego, Paul Albert, N. O’Connor, Kevin McGuinness
- Computer ScienceInternational Joint Conference on Neural Networks…
- 8 August 2019
This work shows that a naive pseudo-labeling overfits to incorrect pseudo-labels due to the so-called confirmation bias and demonstrates that mixup augmentation and setting a minimum number of labeled samples per mini-batch are effective regularization techniques for reducing it.
Bags of Local Convolutional Features for Scalable Instance Search
- Eva Mohedano, A. Salvador, Kevin McGuinness, F. Marqués, N. O’Connor, Xavier Giro-i-Nieto
- Computer ScienceICMR
- 15 April 2016
This work proposes a simple instance retrieval pipeline based on encoding the convolutional features of CNN using the bag of words aggregation scheme (BoW), demonstrating the suitability of the BoW representation based on local CNN features for instance retrieval.
Fully Convolutional Crowd Counting on Highly Congested Scenes
The state-of-the-art for crowd counting in high density scenes is advanced by further exploring the idea of a fully convolutional crowd counting model introduced by (Zhang et al., 2016), and a training set augmentation scheme that minimises redundancy among training samples to improve model generalisation and overall counting performance is developed.
Event detection in field sports video using audio-visual features and a support vector Machine
- D. A. Sadlier, N. O’Connor
- Computer ScienceIEEE Transactions on Circuits and Systems for…
- 1 October 2005
A novel audio-visual feature-based framework for event detection in broadcast video of multiple different field sports and the results suggest that high event retrieval and content rejection statistics are achievable.
Touch Screens for the Older User
It has been 20 years since Ben Schneiderman predicted that there would be an increase in the use of touch screen applications yet it has been only in recent years that this prediction has come to…